Layers of the Atmosphere

Ionosphere Above CA. 82 km

Mesopause (Boundary Layer)
Mesosphere 52-80 km
Stratopause
Stratosphere 14-50 km
Photochemical absorption of ultraviolet light to produce 03 (Ozone).
Tropopause
Troposphere Land/water Surface
To CA. 8km at poles, 16km at equator. Contains CA. 75% of atmospheric mass and
nearly all water vapor, clouds, aerosols.
Boundary Layer - Lowest CA. 1 km
Surface Boundary Layer-Lowest 10m

Layers of the Earth

Lithosphere (100 km thick)

Cool, Rigid
Oceanic Crust (5km)
Continental Crust (30-40 km)
Cryosphere (Ice layer)
Low-velocity Zone
MOHO
Mantle (2885 km thick)
Solid, Rocky
Upper Mantle
Transition (at 400-670 km)
Lower Mantle
D" Layer
Asthenosphere
(600 km thick at 100-700 km)
Hot, Weak,Capable of Flow

Outer Core (2270 km thick)

Molten, Metallic

Inner Core (1216 km thick)

Solid, Iron-rich

(Total Earth Radius 6371 km)

 Geologic Column
( Mesozoic and Cenozoic Eras)

Quaternary Period
Holocene Epoch 12,000
Pleistocene 1.6 Ma
Tertiary
Pliocene
Piacenzian Stage 3,4
Zariclian 5.2
Miocene
Messinian 6.7
Tortonian 10.4
0
ii i Serravallian 14.2
o Langhian 16.3
'o Burdigalian 21.5
IS|
O
c Aquitanian 23.3
a>
O Oligocene
Chattain 29.3
Rupelian 35.4
Eocene (34)
Priabonian 38.6
Bartonian 42.1
Lutetian 50.0
Ypresian 56.5
Paleocene
Thanetian 60.5
Danian 65.0

Cretaceous
Senonian
Maastrichtian 74.0
Campanian 83.0
Santonian 86.6
Coniacian 88.5
o Gallic
UJ Turanian 90.4
lesozoic

Cenomanian 97.0
Albion 112.0
Aptian 124.5
^ Barremian 131.8
Neocomian
Hauterivian 135.0
Valanginian 140.7
Berriasian 145.6
Jurassic
Triassic
LATE CRETACEOUS AND CENOZOIC HISTORY

OF NORTH AMERICAN VEGETATION

CRETACEOUS AND CENOZOIC HISTORY OF

NORTH AMERICAN VEGETATION

NORTH OF MEXICO

ALAN GRAHAM

New York Oxford

Oxford University Press

1999
 Oxford University Press
Oxford New York
Athens Auckland Bangkok Bogota" Buenos Aires Calcutta
Cape Town Chennai Dar es Salaam Delhi Florence Hong Kong Istanbul
Karachi Kuala Lumpur Madrid Melbourne Mexico City Mumbai
Nairobi Paris Sao Paulo Singapore Taipei Tokyo Toronto Warsaw
and associated companies in
Berlin Ibadan

Copyright © 1999 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.
198 Madison Avenue, New York, New York 10016
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.

Graham, Alan.
Late Cretaceous and Cenozoic history of North American vegetation,
north of Mexico / by Alan Graham.
p. cm.
Includes bibliographical references and indexes.
ISBNO-19-511342-X
1. Paleobotany—Cretaceous. 2. Paleobotany—Cenozoic.
3. Plants, Fossil—North America. I. Title.
QE924.G73 1998
561'.197—dc21 97-5713

135798642
Printed in the United States of America
on acid-free paper
 Contents

Prologue vii THREE

Outline vii Context 86
Audiences xii Global Marine Paleotemperature
Acknowledgments xiii Curve 86
References xiii Global Sea-Level Curve 92
Temporal Context 98
Nomenclature xvii Faunas 98
Geoflora Concept 106
ONE Boreotropics Concept 109
Setting the Goal: Modern Vegetation of North America— References 110
Composition and Arrangement of Principal Plant Notes 114
Formations 3
Tundra Formation 3 FOUR
Coniferous Forest Formation 8 Methods, Principles, Strengths, and Limitations 115
Deciduous Forest Formation 16 Paleopalynology (Plant Microfossils) 115
Grassland Formation 18 Paleobotany (Plant Megafossils) 122
Shrubland/Chaparral-Woodland-Savanna References 137
Formation 20 Notes 141
Desert Formation 22
Tropical Elements 25 FIVE
Summary 25 Late Cretaceous through Early Eocene North American
References 25 Vegetational History: 70-50 Ma 142
Notes 27 Context Summary 142
Vegetation Classification 148
TWO Late Cretaceous Vegetation: Maastrichtian,
Cause and Effect: Factors Influencing Composition and 70-65 Ma 150
Distribution of North American Plant Formations through K-T Boundary: 65 Ma 159
Late Cretaceous and Cenozoic Time 28 Origin and Spread of Deciduousness 161
Climate 28 Paleocene Vegetation: 65-56.5 Ma 162
Plate Tectonics 44 Early Eocene Vegetation: 56.5-50 Ma 169
Catastrophes 65 Vegetation Summary 175
References 71 References 177
Notes 85 Notes 186
VI Contents

SIX Vegetation Summary 303

Middle Eocene through Early Miocene North American Notes and References 307
Vegetational History: 50-16.3 Ma 187
Context Summary 187 NINE
Middle through Late Eocene Vegetation: Origins of North American Biogeographic Affinities 318
50-35.4 (34) Ma 191 Floristic Affinities between Southwestern United
Paleoelevation Analysis 207 States-Northern Mexico and Mediterranean
Oligocene Vegetation: 35.4-23.3 Ma 217 Region 319
Early Miocene Vegetation: 23.3-16.3 Ma 223 Floristic Affinities between Warm Deserts of
Vegetation Summary 229 Southwestern United States-Northern Mexico
References 233 and South America 320
Note 241 Floristic Affinities between Eastern North America
and Eastern Asia 321
SEVEN Floristic Affinities between Eastern North America
Middle Miocene through Pliocene North American and Eastern Mexico 324
Vegetational History: 16.3-1.6 Ma 242 Emerging Methods in Biogeography 327
Context Summary 242
Vegetation 246 Epilogue 331
Vegetation Summary 266 References 333
References 268 Notes 336

EIGHT Index 338

Quaternary North American Vegetational History: 1.6 Ma
to the Present 274
Context Summary 274
Pleistocene and Holocene Vegetation 280
 Prologue
The vegetation of North America comprises 17,000-20,000
species of vascular plants or about 7% of the world's flora.
As treated here, these species are arranged into seven plant
formations: tundra, coniferous forest, deciduous forest,
grassland, shrubland/chaparral-woodland-savanna, desert,
and tropical. The current version of these formations is pri-
marily a product of environmental change and biotic re-
sponse during the past 70 million years (m.y.) of the Late
Cretaceous Period and Cenozoic Era. The first unequivocal
angiosperms appear in the Valanginian/Hauterivian [140-
135 millions of years ago (Ma); Crane, 1989,1993; Crane et
al., 1995; Taylor and Hickey, 1996], and by the Late Creta-
ceous they had diversified and radiated to become the
dominant group of terrestrial vascular plants. The Late
Cretaceous also marks the time when most plant fossils
can be assigned to modern forms at least to the level of
family, facilitating the reconstruction of vegetation types,
past environments, migrations, and lineages on the basis
of paleopalynological (plant microfossil) and paleobotani-
cal (plant megafossil) evidence. The development of vege-
tation is a continuum, but for the purposes of this history
the Late Cretaceous is a convenient starting point. The re-
gion under consideration is North America north of Mex-
ico, paralleling the geographic coverage of the Flora of
North America project (Flora of North America Editorial
Committee, 1993).
OUTLINE
The outline used in tracing this history begins with the
composition and arrangement of the modern vegetation
presented in Chapter 1. This sets the goal or end point of
the survey.
In Chapter 2 some of the physical factors are discussed
that have determined terrestrial paleoenvironments and in-
VII
fluenced the course of vegetational history over the past 70
m.y. This material is included because the intent of the sur-
vey is not just to describe stages in the development of
North American vegetation through time, but also to out-
line the principal causes for the changes. A broad overview
of these factors is provided by Crowley and North (1991),
and their general effect on global biotas is summarized by
Zubakov and Borzenkova (1990). There is an excitement
attendant to the realization that we now have the basic
methodologies and are approaching the time when the
myriad processes involved can be integrated into a fuller
outline of the planet's physical and biological history. The
factors are numerous and interactive, however, and until
the development of several large-scale projects and sophis-
ticated data-gathering devices, an explanation of the dy-
namics of environmental change and biotic response was
impossible. These include the Climate Long-Range Investi-
gation and Mapping Program (CLIMAP 1976, 1981, 1984;
which is concerned with reconstruction of the last glacial
and interglacial climate); the Cooperative Holocene Map-
ping Project (COHMAP, 1988; which is concerned with re-
construction of northern hemisphere climates for selected
time intervals of the Holocene), Past Global Changes (PAGES,
a project of the International Geosphere-Biosphere Pro-
gramme, see Workshop Report Series 96-3, 4, 1996), the
Greenland Ice Core Project (GRIP), the Greenland Ice Sheet
Program II (GISP2), the Deep Sea Drilling Project (DSDP),
and its successor the Ocean Drilling Program (ODP); satel-
lites and Geographic Information Systems (GIS); and super-
computers. A convenient way to approach these relation-
ships is through computer modeling, such as the General
Circulation Model (GCM) or Community Climate Model
(CCM) used at the National Center for Atmospheric Re-
search (NCAR), the Geophysical Fluid Dynamics Labora-
tory at the National Oceanographic and Atmospheric Ad-
ministration (GFDL/NOAA), the Goddard Institute for
viii Prologue
Figure 1. Global climate
model grid at 7.8° x 10° res-
olution. Reprinted from
Hansen et al. (1993) with
the permission of the Na-
tional Geographic Society.
Space Studies at the National Aeronautics and Space Ad-
ministration (GISS/NASA), and elsewhere. For those not
familiar with the mathematics and philosophical bases of
computer modeling, some guidelines are useful for assess-
ing the information. General reviews are given in Crowley
and North (1991), Gates (1982), Hannon and Ruth (1994),
Hansen et al. (1993), Imbrie (1982), and Schneider and
Dickinson (1974). More detailed reviews of modeling the-
ory are provided by Schlesinger (1988), Sellers et al.
(1997), Trenberth (1992), and Washington and Parkinson
(1986).
It is important to recognize that modeling presently in-
volves developing, testing, and refining programs that will
eventually approximate real-world situations. For the mo-
ment, however, the programs prescribe certain boundary
conditions (e.g., the position and configuration of conti-
nents at 100 Ma), and hold others quasi-static at modern
values, do not incorporate them, or parameterize (estimate)
them. Then a change called an altered boundary condition
or forcing mechanism is introduced, such as the progres-
sive uplift of a mountain system or an extensive plateau,
and the effects are noted, usually on one to a few factors.
These are called sensitivity tests, in contrast to simulation
models that attempt to reconstruct an actual past climate.
Examples of the latter are rare because all boundary condi-
tions must be specified (e.g., ice, albedo, insolation, CO2
concentration, cloud cover, sea-surface temperatures), and
much of this information is unknown for past times. The
purpose of a sensitivity test may be to determine the effect
of regional orogeny on atmospheric circulation patterns
and subsequent precipitation and temperature regimes.
Often it is found that the results of the simulation (model
performance) correspond to real-world conditions (valida-
tion data or ground truth) based on paleopalynological or
paleobotanical evidence (e.g., development of arid vegeta-
tion to the lee of rising mountains). In other instances the
simulated effect may run counter to known conditions,
prescribing instead high rainfall in areas where arid vege-
tation existed. The first result verifies the accuracy of the
simulation because it replicates a condition known to have
existed. The second result shows that the model needs
improvement.
Model units consist of a series of grids, each represent-
ing areas several hundred kilometers in length and width
(e.g., 7.8° latitude x 10° longitude; Fig. P.I) and a few kilo-
meters in height. The CCM used at NCAR to model North
American climates has a grid resolution of 4.4° x 7.5°. The
geography and continental configuration is idealized and is
commonly called "wonderland" (Fig. P2). The spatial res-
olution or grid constraints of the second simulation noted
above may have allowed the edge of the North American
Cretaceous and early Cenozoic epicontinental sea to ex-
tend too far west on the idealized continent, resulting in
simulated rainfall in an area that, in reality, was dry. This
testing of model prediction on known past climates is
called hindcasting and it gives an estimate of the model's
accuracy for eventual forecasting. Paleontology provides
the principal data base for hindcasting.
The impressive strength of sensitivity tests is that they
can be used to specify and to some extent quantify the
role of individual factors that determine climate. One rea-
son it is not yet possible to completely model climate,
however, is because each of these factors interacts to am-
plify, dampen, or cancel out the effect of the others through
an immensely complex system of feedbacks. On a scale of
geologic time, the position and orientation of the Earth
relative to the sun is changing, solar luminosity varies,
mountains are rising and eroding, seaways are opening and
closing, atmospheric and oceanic circulation patterns are
fluctuating, continents are drifting, volcanoes are erupting,

atmospheric and oceanic CO2 and marine salinity concen-
trations are changing, El Nifios occur, chaos microvibrates
the climate system, glaciers and vegetation types are ex-
panding and contracting, albedoes are changing, catastro-
phes are occurring, and organisms are evolving. The result
is that climatic-biotic interactions are more complex than
our present ability to model them. Nonetheless, modeling
through the use of supercomputers, including parallel
computing with Connection Machines (64,000 simultane-
ous calculations, 10 billion calculations/s), teraflops (com-
puting on the trillionfold/s level), or new photon-based
(rather than electric current-based) microcircuitry allowing
computers to utilize the speed of light, is the only ap-
proach that offers even the possibility of integrating these
factors into a quantified statement of long-term environ-
Prologue ix
Figure 2. Wonderland model
geography. Reprinted from
Hansen et al. (1993) with the
permission of the National
Geographic Society.
mental change. This is especially true if full ocean and at-
mospheric interactions are to be coupled in simulation
models (Bye, 1996).
Chapter 3 discusses results of context information from
other disciplines that is essential for reconstructing past
communities and environments. Two of these, paleotem-
perature and sea-level changes, are also forcing mecha-
nisms (Chapter 2), but because the records are typically
presented as curves that include all of Late Cretaceous and
Cenozoic time, they serve as especially convenient frame-
works for viewing vegetational history. For this reason, pa-
leotemperature and sea-level changes are emphasized sep-
arately in Chapter 3.
A temporal context is provided by radiometric dating.
In addition to the well-known methods based on carbon 14
X Prologue
(14C) and potassium/argon (K/Ar), there are about 15 other
measurements that allow an absolute age to be placed on a
wide variety of sediments and structures, even from
minute samples such as individual seeds or teeth (Saar-
nisto, 1988; Williams et al., 1993, appendix, table A.I).
Among these are rubidium/strontium (Rb/Sr), uranium/
thorium/lead (U/Th/Pb), fission-track dating (Wagner and
Van den Haute, 1992), thermoluminescence dating (Ait-
ken, 1985), electron spin resonance (ESP) dating, amino
acid racemization analysis, and argon 39/argon 40 (39Ar/
40
Ar, including single-crystal laser probe method; Chen et
al., 1996). The application of radiometric techniques facil-
itates correlation between floras in reconstructing regional
vegetation and allows fossil floras from different regions to
be placed in a proper vertical sequence to track environ-
mental, distributional, and lineage changes through time.
They also allow correlation between terrestrial strata and
marine strata that are typically subdivided into nannofossil
and planktic foraminiferal zones. These zones, in turn,
may be aligned with the magnetic polarity time scale (se-
quence of pole reversals), providing a precise chronologi-
cal framework for terrestrial and marine fossiliferous sedi-
ments that lack minerals suitable for radiometric dating.
Seismic profiles through subsurface strata are patterns of
sound reflection that vary with erosion - deposition se-
quences and from which rock types and environments of
deposition can be inferred. These permit a further compar-
ison between terrestrial and marine events (Shackleton
and Opdyke, 1973). The methods are described in numer-
ous publications, including recent ones by Dalrymple
(1991, especially chapter 2), Easterbrook (1988), Good-
friend (1992), McDougall and Harrison (1988), and Roth
and Poty (1989). It should be noted that dates obtained by
the 14C technique and presented in the older literature did
not take into account the changing amounts of radiocarbon
in the atmosphere over time, and they are -(approxi-
mately) 2 ky (thousand years) too young prior to about
12,500 years ago and 250—500 years in maximum error in
later times. For example, the warming period at 14,450 ±
250 years before present (B.P.) is dated at ~12,500 years by
the older 14C measurements. The 14C chronometer has not
been calibrated for dates younger than 40 ka. Unless stated
otherwise, the dates for this interval cited in the text are in
14
C years. Similarly, many of the early dates obtained by
the K/Ar method (Evernden and James, 1964), which was
used to calculate the age of Mesozoic and Tertiary rocks,
were from contaminated sediments and require the conver-
sion of old to new constants (Dalrymple, 1979). Other sam-
ples came from sites adjacent to the ones preserving the
flora or fauna, and some of these were later found to be sig-
nificantly different in age. New and recalibrated dates are
being published, and they are the ones mentioned in the
text.
The framework for describing events in Earth history is
the geologic time scale. This scale continues to be refined;
several versions are available that provide the sequence of
eras, periods, epochs, and stages, as well as estimates of
their span in absolute number of years (e.g., Berggren et al.,
1995; Haq and van Eysinga, 1987; Harland et al., 1989).
The compilations by Berggren et al. (1995) and Harland et
al. (1989) show the age and stratigraphic placement of for-
mations and fossil assemblages mentioned in this text. In
addition to the formal time units shown on the generalized
geologic column in the frontispiece and in the more de-
tailed scales shown in Figs. 5.1, 5.19, 6.13, 6.14, 7.1, and
7.2, other terms encountered in the literature are Cainozoic
(Cenozoic Era), Paleogene or Early Tertiary (Paleocene
through Oligocene), Neogene or Late Tertiary (Miocene and
Pliocene), ice ages (Pleistocene), and Holocene (postglacial
or Recent; time after -12 Kya).
Another important context is faunal history, which re-
flects prevailing vegetation types and general environmen-
tal conditions. Two other contexts are conceptual in nature,
but they nonetheless influence our reading of vegetational
history from the fossil record. These are the older and
much-debated concept of geofloras and the newly emerg-
ing concept of a boreotropical flora.
Other contexts are important to paleoenvironmental re-
constructions, but they are beyond the scope of the present
text. One is the study of land forms and sedimentology
(Ziegler et al., 1987). Tillites, fluvioglacial deposits, ice-
rafted debris, rock striations, and geomorphic features such
as moranes, V-shaped valleys, kettle-hole lakes, aretes, cols,
and cirques all tell of glacial conditions. In contrast, filled
mud cracks, increasing amounts of eolian (wind blown)
dust, and evaporites such as gypsum and salt crystals re-
veal subhumid environments and reduced vegetation cover.
Peat and swamp sediments, which may be subsequently
indurated into lignite and coal, form most extensively
under humid conditions and high water tables in temper-
ate environments and in near-shore coastal marine to brack-
ish waters in tropical environments where biomass produc-
tion exceeds physical, microbial, and oxidative removal.
Bauxites (from Les Baux in Provence, France) are pro-
duced from the weathering of aluminum-containing rocks
under tropical conditions. The characterization of the late
Middle Eocene Green River Flora of Colorado and Utah
as a seasonally arid subtropical shrubland/chaparral-
woodland - savanna is supported by the presence of sea-
sonal evaporites and deeply oxidized redbeds. Another
valuable contribution of sedimentology to paleopalyno-
logical and paleobotanical studies is facies (sediment-type)
analysis. The generalized nature of some fossils (e.g., many
fern spores of the Blechnaceae, Polypodiaceae, Pteridaceae
complex) or their age (pre-Eocene) precludes assigning the
taxa to the proper association and habitat. However, some
of these fossils are consistently found in facies known to be
deposited under specific depositional settings. For exam-
ple, in Paleocene deposits from the Bighorn Basin of
Wyoming, the fern spore Deltoidiospora is most abundant
in sediments representing pond and swamp environments
(Farley, 1989; Farley and Traverse, 1990). Facies analysis

can help reveal the diversity of habitats within a region
and allows some generalized fossil types to be placed in
the proper paleocommunity (Farley, 1990). Even so, it is
usually only the extremes of climate that are revealed,
while within that range there are few clear-cut indications
of climate available from sedimentological analysis alone.
Readers interested in this subject are referred to the stan-
dard texts in sedimentology as they relate to basin analysis
and paleogeography (e.g., Miall, 1984).
Another intent of this text, in addition to providing
cause and effect and context information, is to describe the
impressive array of methodologies available specifically
for the study of fossil floras and to note the strengths and
limitations inherent in each approach. This material is
presented in Chapter 4. The principal disciplines are paleo-
palynology, which for terrestrial vegetation includes the
study of fossil pollen, spores, and phytoliths; and paleo-
botany, which includes the study of plant megafossils such
as leaves, cuticles, cones, flowers, fruits, seeds, and wood.
An important use of the information is in reconstructing
paleoclimates, and two approaches are available for this
purpose.
The modern analog or nearest living relative (NLR)
method involves establishing the composition of fossil flo-
ras by relating various elements to the most morpholog-
ically similar living species. The assemblages are then ar-
ranged in a horizontal sequence to reconstruct regional and
global vegetation and paleoenvironmental conditions at
any one point in time and in a vertical sequence to deter-
mine paleoenvironmental trends, vegetational change, evo-
lutionary patterns, and the origin of biogeographic rela-
tionships on the basis of the ecology and distribution of
presumed modern analogs.
Another approach is foliar physiognomy: the study of
the various leaf size categories, margin types (leaf margin
analysis, LMA), texture variations, and special features
(e.g., the presence or absence of drip tips) found in differ-
ent climatic regimes. In the modern flora the percent repre-
sentation of these features varies in a general way with rain-
fall and especially temperature. The relationship affords an
opportunity to reconstruct paleoclimates independent of
taxonomic identification and the assumed ecological equiv-
alency of fossil and modern species. Methods are being
developed by which qualitative observations and measure-
ments of physiognomic characters can be treated statisti-
cally through multivariate correspondence analysis and
polynomial regressions as part of an important new pro-
gram called the Climate-Leaf Analysis Multivariate Pro-
gram (CLAMP; Wolfe, 1993). The compilation of data for
statistical analysis and the construction of matrices for
computer analyses can expand and enhance the precision
of observations, standardize the information gathered, and
provide a reproducible method of reconstructing paleocli-
mates from fossil leaf assemblages. Ideally, the use of mi-
crofossils and megafossils, and among megafossils the ap-
plication of the modern analog and foliar physiognomy
Prologue xi
approaches, is most effective for paleovegetation analysis.
Where only microfossils or megafossils are available or
where the time interval is beyond the use of modern
analogs (viz., earlier than about the Middle Eocene), recon-
structions are more tentative and depend upon an objec-
tive recognition of the strengths and weaknesses of each
methodology and the available context information.
No single method is adequate for unraveling the com-
plex history of North American vegetation and for tracing
terrestrial paleoenvironments through 70 m.y. of time. In
earlier days it was commonplace for investigators to pur-
sue their speciality independent of others working in the
same discipline and to be mostly indifferent to results from
other lines of inquiry. This frequently led to the existence
of various "schools" (literally and philosophically) where-
from picturesque figures boisterously defended their col-
lections and theories against competitors and new ideas.
They often went to great lengths to establish priority, accu-
mulate authorship of new taxa, increase the holdings of
type specimens for their museums, or to "do a number" on
a colleague who was viewed as particularly obstreperous.
Paleontologists from competing institutions have been
known to write hurriedly scribbled descriptions of new
species on pieces of paper in the field and give them to
couriers who would rush them into town to be published
in the next edition of the local newspaper. In the early
1900s the North American paleontology establishment re-
acted to Wegener's unsettling challenge to the permanence
of continents and ocean basins with a voice that was strong,
virtually united, and ultimately wrong. At the American
Association of Petroleum Geologists' 1926 symposium on
continental drift, University of Chicago geologist Rollin T.
Chamberlain ranted whether we can call geology a science
when it is "possible for a theory such as this to run wild?"
(Sullivan, 1974, p. 11). Paleobotanist Edward W. Berry of
Johns Hopkins University characterized Wegener's method,
and by implication Wegener himself, as not scientific, but
taking "the familiar course of an initial idea, a selective
search through the literature for corroborative evidence, ig-
noring most of the facts that are opposed to the idea, and
ending in a state of auto-intoxication in which the subjec-
tive idea comes to be considered as an objective fact" (Sul-
livan, 1974, p. 12).
It would be naive to assume that such attitudes are ex-
clusively a thing of the past. In addition to valid scientific
debate, the paleontological literature contains its share of
phenomenon chasing, tenacious adherence to previously
expressed opinions, and point-counterpoint arguments
that focus as much on discrediting as on objectively assess-
ing the strengths and weaknesses of a theory. For beginning
students and the general reader, this can result in a confus-
ing melange of opinions expressed as proclamations; over-
emphasized strengths of a particular methodology; and
oversimplistic characterizations of alternative views, ap-
proaches, or theories. Founding practitioners can be prone
to unrestrained advocacy, while the establishment can be
xii Prologue
rather unreceptive to challenge and innovation. Nonetheless,
a hallmark of most present-day investigations is an effort to
place results derived from specialized disciplines into the
broadest possible context based on all available information.
New interdisciplinary coalitions are being forged, and these
are to the benefit of each specialized approach.
With the goal of the survey defined as explaining the
origin and development of the existing plant formations of
North America and discussions of cause and effect, con-
text, and methodologies as background, it is possible to
consider this history in a more integrated way. The record
of North American vegetation is discussed in Chapters 5-8
for the interval between ~70 Ma and the present. Four
stages emerge from three major climatic changes that pro-
duced effects that are evident in the plant fossil record.
Prominent among these are global decreases in tempera-
ture and associated extinction-diversification-migration
events in the biotic realm, which occurred in the Middle
Eocene, the Middle Miocene, and the Quaternary, marking
the beginning of extensive northern hemisphere glacia-
tions. These time segments provide a convenient frame-
work for considering the development of North American
vegetation between the Late Cretaceous and the Early
Eocene (70-50 Ma, Chapter 5), the Middle Eocene and the
Early Miocene (50-16.3 Ma, Chapter 6), the Middle
Miocene and the Pliocene (16.3-1.6 Ma, Chapter 7), and
during the Quaternary (1.6 Ma to the present, Chapter 8).
Emphasis is placed on terrestrial paleoenvironments and
on the time of appearance and subsequent development of
the major plant formations and associations.
Chapter 9 traces the origins of current biogeographic re-
lationships of the North American flora, primarily with the
Mediterranean region, the dry regions of South America,
eastern Asia, and eastern Mexico. Until recently this could
be done through only traditional paleopalynological and
paleobotanical methodologies, estimates of land-sea rela-
tionships through time, and intuitive interpretations based
on the relationships, distributions, and modes of dispersal
of modern taxa. Now it is possible to add new molecular
techniques involving allozyme-isozyme analysis, genetic
distance calculations, and the unsettled issue of the exis-
tence and use of molecular clocks. Patterns of distribution
are being analyzed through emphasis on the movement of
land (vicariance biogeography), as well as through the
movement of organisms (dispersal biogeography). A union
is being attempted between historical geology, biogeogra-
phy, phylogenetics, and cladistics into a synergistic ap-
proach in which results are summarized in the form of area
cladograms. This is also an unsettled methodology, but it
offers the potential of discussing biogeography in ways
other than the traditional narrative format. These innova-
tions constitute a whole new world of research from that
practiced by Linnaeus in the middle 1700s or Asa Gray in
the middle 1800s and their early attempts to define and ex-
plain the origin of geographic affinities of the North Amer-
ican Flora.
AUDIENCES
The audiences to which this text is addressed are those
who share an interest in the environments and events that
have shaped North American vegetation and who have a
general background in the biological and geological sci-
ences. For the specialist, species lists are provided for the
major fossil floras and a list of technical papers is included
after each chapter. For the general reader, terms are defined,
specialized units of measurements are explained, and
widely used common names for familiar plants are given
as they are encountered in the text. At the end of each
chapter a supplemental bibliography of General Readings
is also provided. These include relevant articles in Ameri-
can Scientist, Natural History, Scientific American, Smith-
sonian Magazine, and other sources, as well as reviews in
BioScience, Nature, and Science.
Early versions of the book were also used in a course on
North American vegetation for upper level undergraduates
and graduate students. For this purpose, the presentation
has been designed as a course or seminar text, or as a refer-
ence to be used in conjunction with lectures, student pre-
sentations, and outside readings. In the planning stages it
was anticipated that the subject would be restricted to veg-
etational history and that a few representative floras would
be used to describe North American plant communities at
various times during the Late Cretaceous and Cenozoic.
After using the material in the course, which included stu-
dents with an array of academic backgrounds, it became
clear that a more expansive treatment would serve better.
Many geology students have few or no courses in botany,
while many biologists take only the introductory courses
in geology. Thus, a broader self-contained survey that in-
cluded causes, context, and methodology was deemed most
practical. Three decades of teaching have convinced me
that subjects take on interest in proportion to the back-
ground brought to the course by the students and/or pro-
vided by the instructor, the demonstrated importance or
"sowhatness" of the subject, and the effort spent in pre-
senting and learning the subject. Attempting to provide a
minimum of background and establish the relevancy of the
subject for those mostly having one-time contact with the
topic results in a somewhat encyclopedic text, but the de-
cision was a deliberate and carefully considered one based
on teaching experience.
A perception that has undoubtedly influenced my writ-
ing is that an increasing number of students seem most
comfortable as passive recipients of information. This is in
contrast to learning endeavors that involve active partici-
pation. One of my most challenging and ultimately satisfy-
ing educational experiences as a beginning student was
reading several texts that were important in biology and ge-
ology, but that were outside my immediate speciality and
for which I had only a minimum background. These in-
cluded Carl Swanson's Cytology and Cytogenetics, G. Led-
yard Stebbins, Jr.'s Variation and Evolution in Plants, and

later Verne Grant's Plant Speciation, Robert Sokal and
Peter Sneath's Principles of Numerical Taxonomy, and
Alan Shaw's Time in Stratigraphy. I eventually struggled to
a general understanding and a lasting appreciation of this
material through persistence and a great deal of help from
others. Although the present survey is intended to provide
adequate background for the subject matter discussed, dic-
tionaries, glossaries, additional texts, libraries, peers, and
teachers in related subjects do exist and are sources of ad-
ditional explanatory information. A more full, lasting, and
satisfying understanding of the topic can be gained by as-
suming the role of an active participant and using the text
as part of a cooperative venture.
Terms of nomenclature are explained at the end of the
Prologue and are repeated the first time the symbols are
used. Paleolatitudes and the former positions of continents
follow Scotese and Sager (1989) and Smith et al. (1981).
The species lists given for the principal floras discussed in
the text, together with mention of abundant or otherwise
important members of allied floras, follow the spellings,
family assignments, and nomenclature of the original au-
thor, augmented by recent additions to the flora. A brief de-
scription of families and the general worldwide distribu-
tion of extant families and genera can be found in Willis
(1985); the scientific and common names, range, and diag-
nostic features of North American trees are given in Elias
(1980); distributions and general habitats for extant North
American plant taxa are in the Flora of North America
(Flora of North America Editorial Committee, 1993 et. seq.
1); botanical terms are explained in Allaby (1992); a list of
the common and scientific names of many extinct and ex-
tant mammals of North America is given in Graham and
Mead (1987, appendix I) and in the FAUNMAP Working
Group (1996, footnote 5); and definitions of geologic terms
are given in Allaby and Allaby (1990) and Jackson (1997).
Figures, tables, chapters, and appendices cited in upper
case (e.g., Fig. 1) refer to material in this text, while lower
case (fig. 1) refers to material in other publications.
ACKNOWLEDGMENTS
The genesis of this book was an invitation from Nancy
Morin, Missouri Botanical Garden (presently the American
Association of Botanical Gardens and Arboreta), and Luc
Brouillet, University of Montreal, to prepare a review of
the Late Cretaceous and Tertiary vegetation of North Amer-
ica for the Flora of North America (Graham, 1993). Many
of the illustrations and computer-designed graphics were
prepared by Cathy Brown, William Megenhardt, Larry
Rubens, Diane Sperko, and Samuel Tussing of the AV Ser-
vices at Kent State University. Kirk Jensen and MaryBeth
Branigan at Oxford University Press have skillfully guided
the book to completion.
A number of individuals have read chapters or portions
thereof, or provided discussions, advice, photographs, per-
Prologue xiii
mission to use published and unpublished material, and
other kinds of valuable assistance and cooperation. These
are Theodore Barkely, Kathleen Blase, Larry Bliss, Gerard
Bond, Robert Berner, Thure Cerling, Peter Clark, Rachael
Craig, Peter Crane, Aureal Cross, Margaret Davis, Alastair
Dawson, Hazel Delcourt, Paul Delcourt, David Dilcher, Deb-
orah Elliott-Fisk, David Engebretson, Patrick Fields, Norman
Frederiksen, John Freudenstein, Bruce Graham, Kathryn
Gregory, Eric Grimm, Niels Gundestrup, Jan Hardenbol, Leo
Hickey, Hsioh-Yu Hou, Gordon Jacoby, Herbert Meyer, Karl
Niklas, James MacMahon, Steven Manchester, Rolf Math-
ewes, John Me Andrews, Richard Miller, John Oldow, Do-
lores Piperno, Donald Prothero, William Ruddiman, Sam-
mantha Ruiz, Samuel Savin, Thomas Schmidlin, Linda
Shane, Ruth Stockey, Charles Stockton, Ralph Taggart, Al-
fred Traverse, Bruce Tiffhey, Thomas Van Devender, Terry
Vaughan, S. David Webb, Elisabeth Wheeler, Peter Wigand,
M. A. J. Williams, Scott Wing, Jack Wolfe, and Zhengyi Wu.
Parts of the text were read by several anonymous reviewers
and their suggestions and encouragement were most help-
ful. Introductory chapters were reviewed by biology gradu-
ate student Beverly Brown, and the entire manuscript was
read by Shirley A. Graham. Valued assistance in the tedious
tasks of proofreading and checking literature references was
capably provided by Sonja Kaderly and Earl Landree. The
efforts of these colleagues and students are deeply appreci-
ated. Copyright holders granting permission to use their ma-
terial are acknowledged in the figure legends.
The information summarized here is based on the work
of many individuals. They include early workers who laid
the foundation for Late Cretaceous and Cenozoic paleopa-
lynological and paleobotanical studies in North America.
A new generation of paleopalynologists and paleobotanists
are now refining older concepts and providing new insights
through the innovative application of modern methodolo-
gies. Their contributions are important not only in their
own right, but also in providing an essential part of the
data base required for large-scale environmental modeling
projects and for assessing the real-world accuracy of the
modeled results. In addition, much of what we theorize
about future environmental and evolutionary change must
be assessed against the documented events and trends of
the past. Reseachers in paleontology, including the disci-
plines of paleopalynology and paleobotany, provide a sig-
nificant part of this documentation and their efforts are be-
coming increasingly recognized as necessary for a proper
understanding of the modern Earth and its biota and for
anticipating trends that are likely to shape its future.
References
Aitken, M. J. 1985. Thermoluminescence Dating. Academic
Press, New York.
Allaby, A. and M. Allaby (eds.). 1990. The Concise Oxford
Dictionary of Earth Sciences. Oxford University Press,
Oxford, U.K.
xiv Prologue
Allaby, M. (ed.). 1992. The Concise Oxford Dictionary of
Botany. Oxford University Press, Oxford, U.K.
Berggren, W. A., D. V. Kent, M.-P. Aubry, and J. Hardenbol
(eds.). 1995. Geochronology, Time Scales and Global
Stratigraphic Correlation. Society of Sedimentary Ge-
ology, Tulsa, OK. Special Publ. 54.
Bye, J. A. T. 1996. Coupling ocean-atmosphere models.
Earth-Sci. Rev. 40: 149-162.
Chen, Y., R E. Smith, N. M. Evensen, D. York, and K. R. La-
joie. 1996. The edge of time: dating young volcanic ash
layers with the 40Ar-39Ar laser probe. Science 274:
1176-1178.
CLIMAP Project Members. 1976. The surface of the ice-age
Earth. Science 191: 1131-1137.
. 1981. Seasonal reconstruction of the Earth's surface
at the last glacial maximum. Geological Society of
America, Boulder, CO. Map Chart Series MC-36.
-. 1984. The last interglacial ocean. Quatern. Res. 21:
123-224.
COHMAP Members. 1988. Climatic changes of the last
18,000 years: observations and model simulations.
Science 241: 1043-1052.
Crane, P. R. 1989. Paleobotanical evidence on the early ra-
diation of nonmagnoliid dicotyledons. Plant Syst.
Evol. 162: 165-191.
. 1993. Time for the angiosperms. Nature 366: 631-
632.
, E. M. Friis, and K. R. Pedersen. 1995. The origin
and early diversification of angiosperms. Nature 374:
27-33.
Crowley, T. J. and G. R. North. 1991. Paleoclimatology. Ox-
ford Monographs on Geology and Geophysics. Oxford
University Press, Oxford, U.K. Monograph 16.
Dalrymple, G. B. 1979. Critical tables for conversion of
K-Ar ages from old to new constants. Geology 7: 558-
560.
. 1991. The Age of the Earth. Stanford University
Press, Stanford, CA.
Easterbrook, D. J. (ed.). 1988. Dating Quaternary Sediments.
Geological Society of America, Boulder, CO. Special
Paper 22 7.
Elias, T. S. 1980. The Complete Trees of North America.
Van Nostrand Reinhold, New York.
Evernden, J. E and G. T. James. 1964. Potassium-argon dates
and the Tertiary floras of North America. Am. J. Sci.
262: 945-974.
Farley, M. B. 1989. Palynological facies fossils in nonma-
rine environments in the Paleogene of the Bighorn
Basin. Palaios 4: 565-573.
. 1990. Vegetation distribution across the early
Eocene depositional landscape from palynological
analysis. Palaeogeogr., Palaeoclimatol., Palaeoecol. 79:
11-27.
and A. Traverse. 1990. Usefulness of palyno-
morph concentrations in distinguishing Paleogene
depositional environments in Wyoming (U.S.A.).
Rev. Palaeobot. Palynol. 64: 325-329.
FAUNMAP Working Group. 1996. Spatial response of
mammals to Late Quaternary environmental fluctua-
tions. Science 272: 1601-1606.
Flora of North America Editorial Committee (eds.). 1993.
Flora of North America North of Mexico. Oxford Uni-
versity Press, Oxford, U.K.
Gates, W. L. 1982. Paleoclimatic modeling—a review
with reference to problems and prospects for the
pre-Pleistocene. In: Geophysics Study Committee,
Climate in Earth History. National Academy Press,
Washington, D.C., pp. 26-42.
Goodfriend, G. A. 1992. Rapid racemization of aspartic
acid in mollusc shells and potential for dating over re-
cent centuries. Nature 257: 399-401.
Graham, A. 1993. History of vegetation: Cretaceous (Maas-
trichtian)-Tertiary. In: Flora of North America Edito-
rial Committee (eds.), Flora of North America North of
Mexico. Oxford University Press, Oxford, U.K. Vol. 1,
pp. 57-70.
Graham, R. W. and J. I. Mead. 1987. Environmental fluctu-
ations and evolution of mammalian faunas during the
last deglaciation in North America. In: W. E Ruddiman
and H. E. Wright, Jr. (eds.), North America and Adja-
cent Oceans during the Last Deglaciation. The Geology
of North America, DNAG (Decade of North American
Geology). Geological Society of America, Boulder, CO.
Vol. K-3, pp. 371-402.
Hannon, B. M. and M. Ruth. 1994. Dynamic Modeling.
Springer-Verlag, Berlin.
Hansen, J., A. Lacis, R. Ruedy, M. Sato, and H. Wilson.
1993. How sensitive is the world's climate? Natl.
Geogr. Res. Explor. 9:142-158.
Haq, B. U. and E W. B. van Eysinga. 1987. Geological Time
Table. Elsevier Science Publishers, Amsterdam. Wall
chart.
Harland, W. B., R. L. Armstrong, A. V Cox, L. E. Craig, A. G.
Smith, and D. G. Smith. 1989. A Geologic Time Scale
1989. Cambridge University Press, Cambridge, U.K.
Imbrie, J. 1982. The role of prediction in paleoclimatology.
In: Geophysics Study Committee, Climate in Earth
History. National Academy Press, Washington, D.C.
pp. 21-25.
Jackson, J. A. (ed.). 1997. Glossary of Geology, Amer. Geol.
Inst., Falls Church, VA.
McDougall, I. and T. M. Harrison. 1988. Geochronology
and Thermochronology by the 40Ar/39Ar Method. Ox-
ford University Press, Oxford, U.K.
Miall, A. D. 1984. Principles of Sedimentary Basin Analy-
sis. Springer-Verlag, Berlin.
Roth, E. and B. Poty (eds.). 1989. Nuclear Methods of Dat-
ing. Kluwer Academic Publishers, Dordrecht.
Saarnisto, M. 1988. Time-scales and dating. In: B. Huntley
and T. Webb III (eds.), Vegetation History. Kluwer Aca-
demic Publishers, Dordrecht, pp. 77-112.
Schlesinger, M. E. (ed.). 1988. Physically-Based Modelling
and Simulation of Climate and Climatic Change.
Kluwer Academic Publishers, Dordrecht.
Schneider, S. H. and R. E. Dickinson. 1974. Climate model-
ing. Rev. Geophys. Space Phys. 12: 447-493.
Scotese, C. R. and W. W. Sager (eds.). 1989. Mesozoic and
Cenozoic Plate Reconstructions. Elsevier Science Pub-
lishers, Amsterdam.
Sellers, P. J. et al. (+ 12 authors). 1997. Modeling the ex-
changes of energy, water, and carbon between conti-
nents and the atmosphere. Science 275: 502-509.
Shackleton, N. J. and N. D. Opdyke. 1973. Oxygen isotope
and paleomagnetic stratigraphy of equatorial Pacific
core V28-238: oxygen isotope temperatures and ice
volumes on a 105 and 106 year scale. Quatern. Res. 3:
39-55.
Smith, A. G., A. M. Hurley, and J. C. Briden. 1981. Phanero-

zoic Paleocontinental World Maps. Cambridge Univer-
sity Press, Cambridge, U.K.
Sullivan, W. 1974. Continents in Motion, the New Earth
Debate. McGraw-Hill, New York.
Taylor, D. W. and L. J. Hickey (eds.). 1996. Flowering Plant
Origin, Evolution and Phylogeny. Chapman & Hall,
New York.
Trenberth, K. E. (ed.). 1992. Climate System Modeling.
Cambridge University Press, Cambridge, U.K.
Wagner, G. A. and P. Van den Haute. 1992. Fission-Track
Dating. Enke, Stuttgart, and Kluwer, Norwell, MA.
Washington, W. M. and C. L. Parkinson. 1986. An Introduc-
tion to Three-Dimensional Climate Modeling. Univer-
sity Science Books, Mill Valley, CA.
Williams, M. A. J., D. L. Dunkerley, P. De Deckker, A. P. Ker-
shaw, and T. Stokes. 1993. Quaternary Environments.
Edward Arnold, London.
Willis, J. C. 1985. A Dictionary of the Flowering Plants and
Ferns (revised by H. K. Airy Shaw). 8th (student) edi-
tion. Cambridge University Press, Cambridge, U.K.
Wolfe, J. A. 1993. A method of obtaining climatic parame-
ters from leaf assemblages. U.S. Geol. Surv. Bull. 2040:
1-71.
Ziegler, A. M. et al. (+ 6 authors). 1987. Coal, climate and
terrestrial productivity: the present and Early Creta-
ceous compared. In: A. C. Scott (ed.), Coal and
Coal-Bearing Strata: Recent Advances. Geological So-
ciety of America, Boulder, CO. Special Publ. 32, pp.
25-49.
Zubakov, V. A. and 1.1. Borzenkova. 1990. Global Palaeocli-
mate of the Late Cenozoic. Developments in Palaeon-
tology and Stratigraphy, 12. Elsevier Science Publish-
ers, Amsterdam.
General Readings
Barren, E. J. 1992. Lessons from past climates. Nature 360:
533.
. 1995. Climate models: how reliable are their pre-
dictions? Consequences 1: 17-27.
Bell, G. 1992. Ultracomputers: a teraflop before its time.
Science 256: 64.
Berry, W. B. N. 1987. Growth of a Prehistoric Time Scale,
Based on Organic Evolution. Blackwell Science Pub-
lishers, London. 202 pp.
Broecker, W. 1997. Future directions of paleoclimate re-
search. Quatern. Sci. Rev. 16: 821-825.
Bradley, D. 1993. Will future computers be all wet? Science
259: 890-892.
Cipra, B. 1996. Microprocessors deliver teraflops. Science
271: 598.
Culotta, E. 1993. Is the geological past a key to the (near)
future? Science 259: 906-908.
Emeagwali, P. 1991. The ways of counting. Michigan Today
23: 1-4.
Hanson, B. and D. Voss. 1992. Large scale measurements.
Science 256: 289.
Prologue xv
Kerr, R. A. 1994. Climate modeling's fudge factor comes
under fire. Science 265:1528.
. 1997. Climate-evolution links weaken. Science
276: 1968. [with reference to mammal evolution].
Pool, R. 1992. Massively parallel machines usher in next
level of computing power. Science 256: 50—51.
Trenberth, K. E. 1997. The use and abuse of climate mod-
els. Nature 386: 131-133.
Vrba, E. S., G. H. Denton, T. C. Partridge, and L. H. Burckle
(eds.). 1995. Paleoclimate and Evolution, with Empha-
sis on Human Origins. Yale University Press, New
Haven, CT.
Waldrop, M. M. 1993. Computing at the speed of light. Sci-
ence 259: 456.
References to Frontispiece
Anderson, D. L., T. Tanimoto, and Y.-S. Zhang. 1992. Plate
tectonics and hot spots: the third dimension. Science
256: 1645-1651.
Bloxham, J. and D. Gubbins. 1989. The evolution of the
Earth's magnetic field. Sci. Am. 261: 68-75.
Buffett, B. A., H. E. Huppert, J. R. Lister, and A. W. Woods.
1992. Analytical model for solidification of the Earth's
core. Nature 356: 329-331.
Constable, C. 1993. About turn for reversals. Nature 361:
305-306.
Davies, G. E 1992. Plates and plumes: dynamos of the
Earth's mantle. Science 257: 493-494.
Duba, A. 1992. Earth's core not so hot. Nature 359:
197-198.
Fukao, Y. 1992. Seismic tomogram of the Earth's mantle: ge-
odynamic implications. Science 258: 625-630.
Hoffman, K. A. 1988. Ancient magnetic reversals: clues to
the geodynamo. Sci. Am. 258: 76-83.
Jacobs, J. A. 1992. The Earth's core in a nutshell. Nature
356: 286-287.
Jeanloz, R. 1993. The mantle in sharper focus. Nature 365:
110-111.
Kerr, R. A. 1991. Deep rocks stir the mantle pot. Science
252: 783.
. 1992. Having it both ways in the mantle. Science
258: 1576-1578.
Knittle, E. and R. Jeanloz. 1991. Earth's core-mantle
boundary: results of experiments at high pressures and
temperatures. Science 251: 1438-1443.
Langmuir, C. H. 1993. Deep thoughts on the mantle. Nature
364: 191-192.
Menzies, M. A. (ed.). 1990. Continental Mantle. Clarendon
Press, Oxford, U.K.
Sautter, V, S. E. Haggerty, and S. Field. 1991. Ultradeep
(>300 kilometers) ultramafic xenoliths: petrological
evidence from the transition zone. Science 252: 827-
830.
Vidale, J. E. and H. M. Benz. 1992. Upper-mantle seismic
discontinuities and the thermal structure of subduc-
tion zones. Nature 356: 678-683.
This page intentionally left blank
 Nomenclature

AP arboreal (tree) pollen Kya thousands of years ago

AU Astronomical unit (mean distance between K-T Cretaceous -Tertiary
the Earth and sun; 149,600,000 km, LMA leaf margin analysis
93,000,000 mi) LO last occurrence
B.P. before the present M mountains (mode in computer simulation)
CCM Community Climate Model Ma millions of years ago
chenoam pollen of the families Chenopodiaceae- m.y. million years
Amaranthaceae, which cannot be MART mean annual range of temperature
distinguished MAT mean annual temperature
CLAMP Climate-Leaf Analysis Multivariate Program mb millibar (1/1000 of a bar, or 1000 dynes/cm2)
CLIMAP Climate Long-Range Investigation and MEDD Middle Eocene Diversity Decline
Mapping Program MJ megajoules
COHMAP Cooperative Holocene Mapping Project mt metric tons
D-O Dansgaard-Oeschger event m.y. million years (duration in time)
DSDP Deep Sea Drilling Project NADW North Atlantic deep water
EBM energy balance model NAP nonarboreal (herb) pollen
ENSO El Nino-Southern Oscillation Event NASA National Aeronautics and Space
ESP electron spin resonance (dating) Administration
GCM General Circulation Model NALMA North American Land Mammal Age
GFDL Geophysical Fluid Dynamics Laboratory NCAR National Center for Atmospheric Research
GENESIS global environmental and ecological simula- NEO near Earth object
tion of interactive systems NLR nearest living relative
GIS Geographic Information System NM no mountains (mode in computer
GISP2 Greenland Ice Sheet Program II simulations)
GISS Goddard Institute for Space Studies NOAA National Oceanographic and Atmospheric
GLAS Goddard Laboratory for Atmospheric Sciences Administration
GPTS global polarity time scale ODP Ocean Drilling Program
GRIP Greenland Ice Core Project PAGES Past Global Changes
Gt gigaton (billions of metric tons) PAR pollen accumulation rate
HM half mountains (mode in computer Pg petagram (1015 g = 1 gt)
simulations) pi paleolatitude
ITCZ intertropical convergence zone ppmv parts per million by volume
J joules (energy of 1 amp of current passed RSL relative sea level
through a resistance of 1 ohm for 1 s) SADW south Atlantic deep water
K thousand SCLF single crystal laser fusion
Ky thousand years SCOR scientific committee on oceanic research

XVII
XVIII Nomenclature

SEM scanning electron microscopy TIMS thermal ionization mass spectrometry

SST sea-surface temperature TOMS total ozone mapping spectrometer
TAMS tandem accelerator mass spectrometry VSP vertical seismic profile
TCT pollen of the families Taxodiaceae- WBC western boundary current
Cupressaceae -Taxaceae, which cannot be ~ approximately
distinguished 8 change in (e.g., 818O)
LATE CRETACEOUS AND CENOZOIC HISTORY

OF NORTH AMERICAN VEGETATION

This page intentionally left blank
 ONE
Setting the Goal:
Modern Vegetation of North America
Composition and Arrangement of Principal Plant Formations
Vegetation is the plant cover of a region, which usually
refers to the potential natural vegetation prior to any inten-
sive human disturbance. The description of vegetation for
an extensive area involves the recognition and characteri-
zation of units called formations, which are named with
reference to composition (e.g., coniferous), aspect of habit
(deciduous), distribution (western North America), and cli-
mate, either directly (tropical) or indirectly (tundra). Fur-
ther subdivisions are termed associations or series, such as
the beech-maple association or series within the decidu-
ous forest formation.
Formations and associations constitute a convenient or-
ganizational framework for considering the development
of vegetation through Late Cretaceous and Cenozoic time.
For this purpose seven extant plant formations are recog-
nized for North America (Fig. 1.1, Table 1.1): (1) tundra, (2)
coniferous forest, (3) deciduous forest, (4) grassland, (5)
shrubland/chaparral-woodland-savanna, (6) desert, and
(7) elements of a tropical formation. Several summaries are
available for the modern vegetation of North America, in-
cluding Barbour and Billings (1988), Barbour and Chris-
tensen (1993, vegetation map fig. 5.1), Kuchler (1964), and
Vankat (1979).a The following discussions are based pri-
marily on these surveys.
TUNDRA FORMATION
Tundra (Fig. 1.2) is a treeless vegetation dominated by
shrubs and herbs, and it is characteristic of the cold cli-
mates of polar regions (Arctic tundra) and high-altitude re-
gions (alpine tundra). In the Arctic tundra a few isolated
trees or small stands may occur locally, such as Picea
glauca (white spruce), but these are always in protected
habitats. The Arctic region experiences nearly continuous
darkness in midwinter, and nearly continuous daylight in
3
midsummer. There is a short growing season of only 6-24
weeks; this accounts, in part, for the fact that 98% of all
Arctic tundra plants are perennials (Vankat, 1979). Strong
winds are another feature of the Arctic landscape, often ex-
ceeding 65 km/h for 24 h or more. They likely account for
the frequency of rosettes, persistent dead leaves, and the
cushion growth form, in the center of which wind veloci-
ties may be reduced by 90%. The harsh growing condi-
tions also result in leaves of the microphyllous size class
being comparable to those of desert plants. Vegetative re-
production and self-pollination is common, and pheno-
typic plasticity is high among Arctic tundra plants.
Intensely cold and dry air is maintained by a high-pres-
sure system over the Yukon Territory and the western part
of the Northwest Territories. The importance of this system
in determining the distribution of plant formations in the
north is made evident by comparing the annual variation in
its position to the limits of the Arctic tundra and boreal for-
est. The location of the system varies from a line in winter
that parallels the southern boundary of the boreal con-
iferous forest to a line in the summer that generally paral-
lels the northern boreal coniferous forest-tundra boundary
(Bliss, 1988).
Another feature of the Arctic tundra environment is per-
mafrost, defined as frozen soil in which temperatures per-
sist below 0°C for 2 or more years. Permafrost extends
down 400-600 m in the Arctic region, and it occurs south-
ward in increasingly discontinuous patches into the boreal
forest. There is generally permafrost thawing in summer to
a depth of 20-60 cm and to 100-200 cm along rivers.
The southern boundary of the Arctic tundra is defined
by the limit of permafrost and it follows the contact with
the boreal forest, extending farthest south to ~55°N along
the east coast of North America because of the cold
Labrador Current. The warmer Japan current flows along
the northwest coast of the continent (tundra to ~60°N).
h lteCretaceous and Cenozoic history of Amercican Vegetation
Figure 1.1. Vegetation diagram of North America.
Biodiversity is relatively low in the Arctic tundra; the
North American Arctic consists of -700 species of vascular
plants. Distributions, however, are broad and ~50% of the
vascular plant species are circumboreal. The development
of this formation involved invasion of many preadapted
plants from adjacent alpine (northern and central Rocky
Mountains), coastal, bog, and marsh communities (espe-
cially from montane central Asia), rather than invasion by
a few taxa that subsequently diversified. Evidence for this
are the relatively few genera restricted to tundra habitats.
Rather, most species of tundra plants belong to genera well
represented in other types of environments. The taxonomy
of tundra plants suggests a recent origin for the formation
because of the few genera and the large number of sub-
species and varieties that compose tundra vegetation.
Geographically, the Arctic region is divided into a high
Arctic zone, including the high Arctic polar desert and the
high Arctic semidesert, and a low Arctic zone, which in-
cludes most of the tundra (Table 1.2). The dividing line is
at -72° N latitude. Tundra covers -4 million km2 or -19%
of North America north of Mexico.
The high Arctic polar desert receives less precipitation
than the warm deserts in the southwestern United States;
moisture is available only about 3 months per year. Snow
cover ranges from 20 to 50 cm, and it typically melts by
late July to early August (Bliss, 1988). The low angle of the
sun results in low intensity radiation and cold tempera-
tures, which are usually below 12°C in midsummer. These
environmental factors are reflected in leaf physiognomy;
the leaf area of plants of the high Arctic are generally less
than that for plants of the low Arctic. Vegetation cover
ranges from 0% in areas of permanent snow, to typically
1-5% in ice-free areas with gravelly out-wash soils, to
slightly more in protected areas with better developed
soils. Vast areas of the high Arctic polar desert, including
the Queen Elizabeth Islands, are virtually devoid of plant
cover. Only rosette (e.g., Draba), cushion (Saxifraga), and
mat-forming (Puccinellia) species are able to withstand the
Table 1.1. Summary of North American vegetation types.
Tundra formation Deciduous forest formation associations
High Arctic associations Mixed mesophytic
Graminoid - moss Beech-maple
Dwarf-shrub heath Maple -bass wood
Cushion plant—cryptogam polar semi desert Lake states forest
Cryptogam - herb Oak-chestnut
Graminoid steppe Oak-hickory
Herb barrens Southern mixed hardwood
Low Arctic associations Pine woods
Forest tundra (transitional) Flood-plain forest
Tall shrub
Low shrub Grassland formation associations
Dwarf-shrub heath Tallgrass prairie
Cottongrass-dwarf-shrub heath Mixed grass
Graminoid-moss Shortgrass
Cushion plant-herb-cryptogam polar semidesert Desert grassland
Alpine Palouse Prairie
California grassland
Coniferous forest formation
Boreal coniferous forest (taiga) association Shrubland/chaparral-woodland-savanna formation
Closed forest associations
Lichen woodland Pinon-juniper (and other juniper) woodland
Forest—tundra-ecotone (e.g., aspen-parkland association) Scrub oak—mountain mahogany shrubland
Shrublands (frequently included in Arctic tundra) Chaparral
Bogs Broad-sclerophyll forest
Montane coniferous forest association California woodland
Appalachian montane coniferous forest Blue oak woodland
Western montane coniferous forest Southern oak woodland
Rocky Mountains region Mixed evergreen forest
Riparian and canyon forest Savanna
Short conifer woodland
Pinus ponderosa woodland Desert formation associations
Madrean pine-oak woodland Great Basin
Pseudotsuga menziesii forest Mojave Desert
Cascadian forests Sonoran Desert
Montane serai forests Chihuahuan Desert
Subalpine white pine forests Tropical and subtropical elements
Treeline vegetation
Meadows and parks
Sierra Nevada region zone forests
Montane (transition)
Upper montane (Canadian)
Subalpine (Hudsonian)
Pacific Northwest region
Pseudotsuga menziesii-Tsuga heterophylla forest
Picea sitchensis-Tsuga heterophylla forest
Sequoia sempervirens forest
Klamath Mountains mixed evergreen forest
Sierran-type mixed conifer forest
Subalpine forests and parklands
Abies amabilis-Tsuga heterophylla forest
Abies magnifica var. shastensis forest
Tsuga mertensiana forest
Subalpine parkland

Based on Harbour and Billings (1988), Barbour and Christensen (1993), and Vankat (1979).
 Figure 1.2. Tundra
formation—Graminoid-moss
association (Carex stans,
Eriophorum angustifolium],
Devon Island, Northwest
Territories. Photograph
courtesy of Larry Bliss.

Table 1.2. Comparison of environmental and biotic characteristics of low and high Arctic in North America.
Characteristics Low Arctic High Arctic
Length of growing season (months) 3-4 1.5-2.5
Mean July temperature 8-12/35.6 3-6/20.8
(°C)/ppt (mm)
Frobisher Bay, NWT, 64° 3.9/53
Baker Lake, NWT, 65° 10.8/36
Tuktoyaktuk, NWT, 70° 10.3/22
Kotzebue, AK, 67° 12.4/44
Umiat, AK, 69° 11.7/24
Anamagssalik, Greenland, 65° 7.4/35
Godthaab, Greenland, 64° 7.6/59
Umanak, Greenland, 71° 7.8/12
Barter Island, AK, 70° 4.6/18
Barrow, AK, 71° 3.8/22
Cambridge Bay, NWT, 69° 8.1/22
Sacks Harbour, NWT, 72° 5.5/18
Resolute, NWT, 75° 4.3/26
Isachsen, NWT, 79° 3.3/21
Eureka, NWT, 80° 5.3/13
Alert, NWT, 80° 3.9/18
Scoresbysund, Greenland, 70° 4.7/38
Nord, Greenland, 81° 4.2/12
B otanical/vegetational
Total plant cover (%)
Tundra 80-100 80-100
Polar semidesert 20-80 20-80
Polar desert 1-5 1-5
Vascular plant flora (species) 700 350
Bryophytes Common Abundant
Lichens (common growth form) Fruticose, foliose Crustose, foliose
Vascular (common growth form) Woody, graminoid, Graminoid, cushion, rosette
Precipitation (ppt). Adapted from Bliss (1988).

extreme climate. In one survey of six islands, only 17
species of vascular plants and 14 species of cryptogams
were recorded. The vascular plant cover averaged 1.8%,
the cryptogams 0.7%, and 98% of the ice-free area was
bare soil. The high Arctic semidesert is often included
within the polar desert. It covers ~25% of the northern is-
lands and 50-55% of the southern islands. The vegetation
consists of cushion plant-cryptogam and cryptogam-herb
species. Compared to the low Arctic tundra, the high Arc-
tic tundra generally consists of herbaceous rather than
woody species; there is much less plant cover; and lichens
and mosses are much more prominent (up to 95% of the
ground cover in some areas) with graminoid-moss tundra
also extensively developed.
The low Arctic region extends from the high Arctic zone
south to the northern limits of the boreal forest. It is char-
acterized by a snow cover of 1-5 m or more that usually
does not melt until about mid-June. The plant cover is
nearly continuous, consisting of tall shrubs (2-5 m) to
dwarf shrub heath (5-20 cm), with abundant grasses,
sedges, lichens, and mosses. There is a general increase
southward in total plant cover, plant height, and woody
vascular plant species and a decrease in the prominence of
lichens and mosses, although they are still a conspicuous
and important component of the vegetation. The low Arc-
tic tundra is more diverse in composition than that of the
high Arctic and will be used to describe the vegetation in
more detail.
Six principal types of low Arctic tundra are recognized
(Bliss, 1988). The tall shrub association has Salix alaxensis
(willow) as a common member, and also includes S. glauca
spp. richardsonii, S. pulchra, and Alnus crispa (alder).
Species of Betula (birch) grow along river terraces, stream
banks, steep slopes, and lake shores where they form the
tallest component of the vegetation (2-5 m in height).
The low shrub association consists of plants generally
between 40 and 60 cm in height. The prominent ones are
Betula nana ssp. exilis, and Salix glauca spp. richard-
sonii, with a ground cover of Carex lugens (sedge), C.
bigelowii, Eriophorum vaginatum (cottongrass), Vac-
cinium uliginosum (heath), V. vitis-idaea spp. minus, Em-
petrum nigrum ssp. hermaphroditum, Ledum palustre
spp. decumbens, Arctostaphylos alpina, A. rubra, and
Rubus chamaemorus. The common lichens are Cetraria
nivalis, C. cucullata, Cladonia gracilis, C. mitis, C. rangife-
rina, and Thamnolia vermicularis. The mosses include
Aulacomnium turgidum, Hylocomium splendens, and Poly-
trichum juniperinum.
The dwarf-shrub heath association is usually found in
relatively small parcels of a few hundred square meters. In
North America it occurs scattered from Alaska to Hudson
Bay and into Greenland on well-drained soils of river ter-
races, slopes, and uplands where winter snows are at least
20-30 cm deep. The vegetation includes Ledum palustre
spp. decumbens, Vaccinium uliginosum, V. vitis-idaea, Em-
petrum nigrum spp. hermaphroditum, Loiseleuria procum-
Setting the Goal: Modern Vegetation of North America 7
bens, Rhododendron lapponicum, R. hamschaticum, Cas-
siope tetragona, Betula nana spp. exilis, B. glandulosa,
deformed Picea mariana (black spruce), dwarf Salix reticu-
lata and S. herbacea, Dryas integrifolia, Phyllodoce caer-
ulea, Carex bigelowii, C. misandra, Luzula nivalis, and var-
ious lichens and mosses.
The cottongrass-dwarf shrub heath association is com-
mon along rolling uplands between mountains and along
the wet coastal plain of Alaska, the Yukon Territory, and
the Mackenzie River delta. The conspicuous element in
this association is the cottongrass Eriophorum vaginatum
spp. vaginatum in the west and E. vaginatum spp. spissum
in the east. Heath shrubs include Vaccinium vitis-idaea
spp. minus, V. uliginosum spp. alpinum, Ledum palustre
spp. decumbens, and Empetrum nigrum spp. hermaphro-
ditum. Other shrubs are Betula nana spp. exilis and Salix
pulchra. The ground cover consists of two common sedges,
Carex bigelowii and C. lugens; the mosses Dicranum elon-
gatum, Aulacomnium turgidum, A. palustre, Rhacomi-
trium lanuginosum, Hylocomium splendens, Tomenthyp-
num nitens, and several species of Sphagnum; and lichens
such as Cetraria cucullata, C. nivalis, Cladonia rangiferina,
C. mitis, C. sylvatica, Dactylina arctica, and Thamnolia
vermicularis.
The graminoid-moss association is a treeless wetland
with grasses and sedges as the prominent elements. The as-
sociation is common along foothills and coastal regions in
Alaska. The sedges are Carex aquatilis, C. rariflora, C. ro-
tundata, C. membranaceae, Eriophorum angustifolium, E.
scheuchzeri, and E. russeolum; grasses are Arctagrostis lat-
ifolia, Dupontia fisheri, Alopecurus alpinus, and Arcto-
phila fulva; mosses are Hylocomium splendens, Tomen-
thypnum nitens, and species of Aulacomnium, Ditrichum,
Calliergon, Drepanocladus, and Sphagnum. Other associ-
ates include Menyanthes trifoliata, Equisetum variegatum,
Arctophila fulva, Potentilla palustris, and Hippuris vul-
garis. This type of tundra extends eastward, but it is less
common in Greenland. In more temperate regions grasses
often form salt marshes and true grasslands, but these asso-
ciations are rare in the Arctic.
The cushion plant-herb-cryptogam polar semidesert
association occurs on wind-swept slopes and ridges,
which are often dominated by cushions or mats of Dryas
integrifolia and D. octopetala. These species may be a com-
ponent of fellfields, which are rocky areas in cold climates
with shrubs growing between the rocks and boulders. They
are most common in the extensive areas of barren rock and
gravel of the eastern Arctic. Other cushion plants are Si-
lene acaulis, Saxifraga oppositifolia, and S. tricuspidata
associated with the lichens Alectoria nitudula, A. ochro-
leuca, Cetraria cucullata, and C. nivalis, and the moss Rha-
comitrium lanuginosum.
Alpine tundra occurs in the Cascade Mountains, the
Sierra Nevada, the Rocky Mountains, and the Appalachian
Mountains of central New England northward at elevations
above -1200 m. Compared to the low Arctic zone, alpine
8 Late Cretaceous and Cenozoic History of North American Vegetation
regions above timberline have more uniform day and night
periods throughout the year, light intensity is greater, and
there is greater fluctuation in temperatures. For example,
temperatures may rise to 40°C at the soil surface during
summer days and may drop by 20°C in a few seconds
under passing clouds. The soil is generally drier in an
alpine tundra due to faster drainage on the steep slopes.
The substrate is often rocky, and the plants grow among the
rocks and boulders to form fellfields. Most alpine tundras
include wet meadows dominated by sedges and grasses
and dry meadows that are drained by the later part of the
summer. Tall shrub communities of Salix may occur along
streams.
Many alpine tundra plants are also found in the Arctic
tundra, suggesting a common origin. For example, 45% of
the plants in an alpine tundra from the Rocky Mountains
in Montana and 75% in an Appalachian Mountain tundra
in New Hampshire also occur in the Arctic tundra (Vankat,
1979).
CONIFEROUS FOREST FORMATION
Boreal Coniferous Forest or Taiga
One of the defining and most obvious features of the boreal
coniferous forest is the predominance of a relatively few
genera of gymnosperms (e.g., Abies, Larix, Picea, Thuja, a
few species of Pinus}. Gymnosperms are also dominant
along the Gulf Coast of the southeastern United States and
on the higher mountain slopes of the American west. In my
view there is probably a historical component to the expla-
nation for the abundance of gymnosperms in these regions.
Each of the areas is characterized by physiological aridity,
even though precipitation may be moderate or even high.
In the boreal coniferous forest, soil water is frozen for 8—9
months of the year. Along the Gulf Coast of the southeast-
ern United States the soils are predominantly deep sand
where water percolates through rapidly; at mid- to high el-
evations in the west, surface runoff from the mountain
slopes reduces opportunity for absorption. The origin and
early diversification of modern gymnosperms was primar-
ily during the climatically tumultuous times of the Per-
mian, which included widespread aridity as reflected by
evaporites and extensive, deeply oxidized redbeds. As a
consequence, gymnosperms possess a number of features
that allow them to cope with climatic and physiological
aridity and to occupy underexploited ecological niches
such as regions of frozen soil, deep sand, and steep slopes.
Adaptive morphological features include sunken stomata,
thick cuticles, circular leaves (or circular in the aggregate
fasicles) that present minimum surface area for water loss,
and convoluted mesophyll cells that provide increased
surface area to maintain adequate physiological activity in
compensation for reduced leaf volume. Physiological
adaptations include tannins stored in the cell vacuoles that
constitute a colloid system. As water is lost from the
leaves, a point is reached where the vacuole contents
change from a liquid to a gel, further reducing water loss.
One environment that renders plants with these adapta-
tions at a competitive advantage is in the region just below
the low Arctic tundra.
The boreal coniferous forest (Fig. 1.3; Shugart et al.,
1992) covers ~28% of North America north of Mexico and
extends from the low Arctic tundra to transition zones
with various vegetation types to the south.2 In eastern
North America it intergrades with the hemlock and white
pine-northern hardwood forest association (lakes state
forest) of the deciduous forest, while in the midcontinent
region there is a transition through a deciduous forest
phase into the central prairie. In the west it merges with
the Rocky Mountains and northwest coastal coniferous for-
est, sharing such species as Picea glauca, Pinus contorta
(lodgepole pine), and Populus tremuloides (quaking aspen).
The ecotone with the low Arctic tundra is comparatively
sharp and narrow in the east and west, but it is up to 235
km wide in central Canada (Elliott-Fisk, 1988).
Environments are generally similar and grade northward
into those of the low Arctic tundra: low temperatures, low
precipitation, and a short growing season of 75-120 days
(Table 1.3). The evergreen habit of the dominant trees is a
response to the short growing season. Temperatures range
from a summer maximum of 15 °C to a winter minimum of
-35°C. The distribution is determined by a complex of fac-
tors, but temperature is significant. In general, the northern
limit of the boreal coniferous forest follows the 13°C July
isotherm, while the southern limit corresponds to about the
18°C July isotherm (Elliott-Fisk, 1988). Rainfall within the
formation ranges from about 240-500 mm, but water is
frozen for 8-9 months of the year.
The area was extensively glaciated in the east by the
Laurentide ice sheet and in the west by the Cordilleran ice
sheet, resulting in a flat topography with numerous lakes.
Typical succession on these lakes is from open water to en-
croaching Sphagnum mat, shrubs, trees, and finally to a
filled basin covered with boreal forest trees. The accumu-
lating organic material produces poorly oxygenated, acidic
environments that reduce microbial activity and favor
preservation of pollen, spores, seeds, wood, and cuticles,
as well as diatoms, phytoliths, insects and other inverte-
brates, and even small mammals. As a consequence, a great
deal is known about the late-glacial and postglacial history
of the boreal forest from pollen, spore, and other analyses.
Fire is an important factor in maintaining the coniferous
forest (Crutzen and Goldammer, 1993; Johnson, 1992).
Many tree species regenerate quickly after fire by sprout-
ing, and fire is important in seed dispersal of serotinous -
cone species. These cones are sealed by resin that melts
during a fire, releasing the seeds to a cleared, more nutri-
ent-rich, less competitive environment. In the past, fire
was a frequent and natural component of the coniferous
forest environment, resulting from volcanic activity, light-
 Figure 1.3. Coniferous forest formation—boreal coniferous forest (taiga) association (Abies balsamea, Larix laricina,
Picea glauca, P. mariana), Chippewa National Forest, Minnesota. Photograph courtesy of the National Agricultural
Library, Forest Service Photo Collection.

Table 1.3. Climatological data for northern boreal forest sites in Alaska and western, central, and eastern Canada.
Parameter Fairbanks, Alaska Doll Creek, Yukon Ennadai Lake, NWT Napaktok Bay, Labr.
Mean daily temp. (°C)
(January) -24 -35 -31.5 -20
(July) 15 14 13 7.5
Mean annual temp. (°C) -3 -10 -9.3 -5
Mean annual precipitation (mm) 290 240 290 500
Mean growing season length (days) 100 120 100 75
Mean annual net radiation 22 20 19 18
(kly/yr)*
Adapted from Elliot-Fisk (1988).
* Kilolangley/year.
10 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 1.4. Coniferous forest formation—boreal coniferous forest association (Pinus

banksiana), Chippewa National Forest, Minnesota. Photograph courtesy of the National
Agricultural Library, Forest Service Photo Collection.

ning, and in more recent times from the hunting and agri-
cultural practices of the Amerindian. The ground was
cleared of debris and the fires scorched only the lower,
older trunks of the trees, facilitating seed dispersal and
sprouting. In recent times management policy has been
largely to prevent fires, with the result that underbrush ac-
cumulates in great abundance. Fires now reach the crowns
and frequently cause much more extensive and long-term
damage. Another modern factor significantly affecting the
range of the coniferous forest is acid rain, defined as pre-
cipitation with a pH of less than 5.6. There are areas in the
northeastern United States where the pH of rain water is as
low as 3.0.
The trees of the boreal coniferous forest are compara-
tively small, usually under 15 m in height. At its northern
limit approaching the low Arctic tundra, the trees are re-
duced to ~4 m in height or to even smaller shrubby forms,
the canopy becomes more open, and the forest breaks up
into small stands or individual trees.
The closed forest phase of the boreal coniferous forest
(Fig. 1.3) has the lowest species diversity of any North
American forest vegetation, with only nine principal tree
genera; only the tundra is lower in number of species. This
is likely due to disturbances during the Quaternary and to
the youthfulness of much of the landscape. Picea glauca is
the most widespread tree and is common on acidic, allu-
vial soils along river valleys. Picea mariana is the most fre-
quently occurring tree in the boreal forest. It is common on
upland, basic, shallow soils or bedrock, and on poorly
drained, acidic soils surrounding Sphagnum bogs. Other
trees include Abies balsamea (balsam fir), Larix laricina
(tamarack, larch), Pinus banksiana (jack pine; Fig. 1.4), P.
contorta, Thuja occidentalis (white cedar), Betula pa-
pyrifera (paper buch). populus balsamifera (balasam
poplar), P. tremuloides, and coastal species such as Tsuga
heterophylla and Chamaecyparis nootkatensis (Alaskan
cedar and yellow cypress). Quaking aspen is dominant in
the aspen-parkland association that forms one transition
between the boreal coniferous forest and the northern part
of the grassland formation. The aspen-parkland association
is often serai (temporal), as evidenced by the frequent uni-
form size of the trees in individual stands. Shrubs include
Alnus crispa (green alder), Betula glandulosa (dwarf
birch), Salix sp., and various Ericaceae (heaths), although
these are mostly part of the low Arctic tundra. There is an
understory of lichens (Cladonia or reindeer lichen), mosses
(Hylocomium splendens, Pleurozium schreberi, and espe-
cially Sphagnum sp.), ferns (Osmunda), and lycopods (Ly-
copodium). The northern region of the forest is abundant
in lichens and is often treated as a separate phase called
the lichen woodland. It includes Picea mariana, P. glauca,
and the lichens Stereocaulon paschale in the west and
Cladonia stellaris in the east.
The bog flora of the boreal coniferous forest includes
Sphagnum and trees and shrubs on or surrounding the
basin, depending on the state of bog maturity and the ex-
tent of basin filling. The common trees are Picea mariana
and Larix laricina that are associated with orchids, heaths,
composites, sedges, and grasses. In southern Alaska Tsuga
heterophylla, Thuja plicata (western arborvitae), Chamae-
cyparis nootkatensis, and Pinus contorta generally replace
Picea mariana.

Montane Coniferous Forest
Appalachian Montane Coniferous Forest
This forest is a continuation of the boreal coniferous forest
formation, and it is generally similar in composition from
north to south (Greller, 1988). Toward the south it occurs at
progressively higher elevations: above 150 m in Maine, 762
m in the Green Mountains of Vermont and in the White
Mountains of New Hampshire, 1280 m in the Catskill Moun-
tains of southeastern New York state, and 1524 m in the
Great Smoky Mountains of Tennessee and North Carolina
[e.g., Mt. Mitchell, 2037 m (6684 ft), and Clingmans Dome,
2025 m (6643 ft), the highest peaks east of the Rocky
Mountains]. The average temperature is ~5°C for the cold-
est month and ~15°C for the warmest month. Annual pre-
cipitation is 1500 mm in the northern part of the range and
2250 mm in the southern part. Thus, the Appalachian mon-
tane coniferous forest occupies regions that are warmer and
wetter than the boreal coniferous forest.
The dominant trees are Abies at the higher elevations
with Picea intermixed at lower elevations. The most im-
portant fir in the north is A. balsamea; in the south it is re-
placed by A. fraseri (fraser fir), a closely related species
that may have evolved via founder effect mechanisms in
relatively recent times. The principal spruce throughout is
Picea rubens (red spruce). Associated with the coniferous
element is Betula papyrifera in the north, B. lutea (yellow
birch) throughout the range, and Rhododendron (azalea,
rhododendron) on exposed slopes in the south.
Western Montane Coniferous Forest
Rocky Mountains Region The Rocky Mountains are the
most continental of the western mountain chains. The pre-
vailing westerlies provide the highest annual precipitation
on the western slopes, ranging from 400 mm in the south to
1500 mm in the north. The coniferous forest of the Rocky
Mountains extends from northern Alberta to New Mexico,
with outliers at higher elevations from central Utah to
western South Dakota (e.g., Picea glauca in the Black
Hills). To the west, coniferous forests of the Sierra Nevada
and the Cascade Mountains and Coast Ranges of the Pacific
Northwest are generally recognized as separate floristic
provinces. In the north the Rocky Mountains coniferous
forest blends into the boreal coniferous forest, and in the
south at lower elevations it frequently borders woodland
vegetation of Pinus edulis (pirlon pine) and various species
of Juniperus (juniper). It also extends along the Sierra
Madre Oriental and Sierra Madre Occidental to the Trans-
volcanic Belt of Mexico (the Madrean region of Axelrod
and Raven, 1985), giving the formation a latitudinal range
in the New World of from -65° N to -19° N latitude.
Zonation is one of the most evident and widely recog-
nized aspects of the Rocky Mountains vegetation, espe-
cially since the definitive work of Merriam (1890) describ-
Setting the Goal: Modern Vegetation of North America 11
ing six altitudinally arranged life zones. For the region of
northern Arizona and adjacent areas these zones, as classi-
cally described, are the lower Sonoran life zone (a southern
desert scrub, to -200 m elevation), upper Sonoran life zone
(desert grassland to 1100 m) and pinon and juniper wood-
land (to 1700 m), transition life zone (ponderosa pine for-
est to 2100 m), Canadian life zone (douglas fir forest to
3000 m), Hudsonian life zone (spruce and fir forest to 3500
m), and Arctic and Alpine life zone (alpine tundra to 3800
m; Lowe, 1980). Three of these (transition, Canadian, and
Hudsonian) are characterized by coniferous forest. This life
zone nomenclature has been modified in recent times, es-
pecially in regions beyond the deserts where the altitudi-
nal belts are commonly referred to as foothill (transition),
montane (Canadian), subalpine (Hudsonian), and alpine
(Arctic and Alpine).
Peet (1988) notes that, "Given the length of the cor-
dillera, the surprise is not that latitudinal variation exists
in Rocky Mountain forests but that major vegetation types,
often representing elevational zones, remain essentially
constant over great distances." Elevation, moisture gradi-
ents, and soil combine to produce the recognizable units
that do exist, and Peet (1988) recognizes 11 vegetation
types.
Riparian and canyon forest in the drier lowlands consist
ofPopulus trichocarpa, P. sargentii, P. fremontii, P. angus-
tifolia (cottonwoods), and Salix sp. With increasing eleva-
tion these species are replaced by P. angustifolia, Alnus
tenuifolia, and Betula occidentalis; to the south bordering
the Madrean region Platanus wrightii (sycamore), Juglans
major (walnut), Fraxinus velutina (ash), and Alnus ob-
longifolia become the common deciduous species.
A short conifer woodland of Juniperus monosperma
(one-seed juniper) and Pinus edulis forms a transition be-
tween the drier plains vegetation and the montane conifer-
ous forest of the southern Rocky Mountains. At higher alti-
tudes from Mexico to southern British Columbia, the lower
montane zone is characterized by a Pinus ponderosa wood-
land, although overgrazing has caused species of Quercus
(oak) and Cercocarpus (mountain mahogany) to replace
ponderosa pine in many areas. The effect of overgrazing on
ponderosa pine may be indirect in that when grasses are
reduced, fewer fires control the competing shrubby vegeta-
tion. The southern margin of the ponderosa pine associa-
tion grades into the Madrean pine-oak woodland of the
southwestern United States and Mexico.
Near the upper elevational limits of the ponderosa pine,
the Pseudotsuga menziesii forest forms a transition to the
upper montane zone. Peet (1988) recognizes a Cascadian for-
est and a montane serai forest. The latter is a consequence
of cyclic and unpredictable disturbances that are charac-
teristic of the elevated and tectonically active landscape.
Even though the communities are theoretically serai in na-
ture, the frequency of disturbances such as fire, wind, in-
sects, disease, browsing, avalanches, landslides, extreme
weather, volcanism, and human activity make these com-
12 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 1.5. Coniferous forest formation—montane coniferous forest association (treeline vegetation; Abies la-
siocarpa, Picea engelmanii), Colorado.
munities a constant, albiet shifting, feature of the Rocky
Mountains vegetation. Two common members of the serai
forest are Populus tremuloides, the most widespread tree
in North America, and Pinus contorta.
Members of the white pine group (Pinus subgenus Hap-
loxylon) are able to withstand the low temperatures and
strong winds of high-altitude ridges and exposed slopes, so
they form subalpine white pine forests. The important
species are P. albicaulis (whitebark pine) in the north, P.
aristata (bristlecone pine) in the south, and P. flexilis (lim-
ber pine) throughout most of the region. Pinus longaeva
(Great Basin bristlecone pine) replaces bristlecone pine on
the high peaks of the Great Basin. Treeline vegetation (Fig.
1.5) typically consists of Picea engelmannii (Engelmann
spruce) and Abies lasiocarpa (subalpine fir); members of
the white pine group grow on exposed sites.
Open areas of mostly grasses and sedges occur scattered
through the coniferous forest of the Rocky Mountains. The
smaller ones are called meadows and the larger ones parks.
The presence of meadows and parks at lower elevations
(e.g., in valleys like Estes Park, CO) is thought to result
from a combination of soil texture (typically fine alluvial
soils), drought, and exposure on south-facing slopes and
steep hillsides. At high elevations, saturated peaty soils,
low temperatures, and high winds are considered to be im-
portant factors. At midaltitude these communities include
Carex utriculata, C. aquatilis, and Calamagrostis canaden-
sis; at higher altitudes Eleocharis pauciflora, Caltha lep-
tosepala, Deschampsis caespitosa, Carex illota, Poda-
grostis humilis, Phleum alpinum, Poa reflexa, Erigeron
peregrinus, and Bistorta bistortoides are common compo-
nents. Occasional boggy areas include Sphagnum and the
ericads Vaccinium myrtillus, Kalmia polifolia, and Gaul-
theria humifusa.
Sierra Nevada Region Altitudinal zonation is evident
among the plant communities of the Sierra Nevada, but the
proximity to the Pacific Ocean causes the ecotones to be
higher. For example, in the southern Sierra Nevada alpine
tundra begins at -3500 m, while in the Cascade Mountains
and mountains further north it begins at -2000 m. In the
south the area is characterized by cool wet winters and dry
warm summers (Mediterranean climate), and 80-85% of
the precipitation falls during the winter months. Farther
north winter precipitation constitutes 60-65% of the an-
nual total, mostly as snow, which may reach 20 m annually
in the subalpine zone. Along the western slopes of the
Sierra Nevada rainfall averages 250-350 mm at low eleva-
tions and 1250 mm in the upper montane and subalpine
zones.
The montane (transition) zone forest of the Sierra Nevada
(Fig. 1.6) typically begins at -1500 m and extends to -2500
m. It includes Pinus ponderosa as the dominant element,
with P. jeffreyi (Jeffrey pine) growing at higher elevations;
both extend into the upper montane zone. Pinus lamber-
tiana (sugar pine) is present in the central Sierra Nevada,

Figure 1.6. Coniferous forest formation—montane coniferous forest association (Sierra Nevada region,
montane zone forest; Abies concolor}, California. Photograph courtesy of the National Agricultural Library,
Forest Service Photo Collection.
and Sequoiadendron giganteum (giant sequoia; Fig. 1.7) is
dominant in scattered groves in the central and southern
Sierra Nevada at elevations from 825 to 2680 m. Other conif-
erous elements, often representing extensions from montane
communities, are Pseudotsuga menziesii, Abies concolor
var. lowiana (California white fir), and Calocedrus decurrens
(= Libocedrus decurrens, incense cedar). Associated an-
giosperm trees are Quercus kelloggii (black oak), Q. chryso-
lepis (canyon live oak), Cornus nuttallii (dogwood), Acer
macrophyllum (bigleaf maple), and Cercocarpus ledifolius;
associated shrubs are Arctostaphylos, Ceanothus, Chamae-
batia, Castanopsis, Lithocarpus, Prunus, Ribes, Symphori-
carpus, and Vaccinium (Barbour, 1988).
The upper montane (Canadian) zone forest extends
from 2500 to 3000 m and is characterized by Pinus con-
torta var. murrayana, which is occasionally replaced at
higher altitudes on mesic sites by Abies magnified (red fir).
Serai stands ofPopulus tremuloides form local aspen park-
lands (Barbour, 1988). Understory shrubs are not common,
but Arctostaphlos, Ceanothus, Castanopsis, and Ribes can
be found.
The subalpine (Hudsonian) zone forest extends typi-
cally from 3000 to 3500 m. These forests differ from those
in the Rocky Mountains in that they are not characterized
by Picea and Abies. Rather, the dominant species are Pinus
albicaulis, P. flexilis, Tsuga mertensiana (mountain hem-
lock), Pinus contorta var. murrayana, P. monticola (west-
ern white pine), and P. balfouriana spp. austrina (foxtail
Setting the Goal: Modem Vegetation of North America 13
pine). Associates include Populus tremuloides var. aurea,
Pinus Jeffrey!, P. monophylla, Abies concolor var. lowiana,
A. magnifica, Juniperus occidentalis spp. australis, and oc-
casional prostrate shrubs of Arctostaphylos, Artemisia,
Holodiscus, Phyllodoce, Salix, Ribes, and Vaccinium.
Along the colder and drier eastern slopes, altitudinal
zones are not as distinct. The upper montane zone (2500-
2900 m in the northern Sierra Nevada, 2800-3400 m in the
south) includes Abies magnifica, A. concolor var. lowiana,
and Pinus contorta; the lower montane zone (2000-2500
m in the north, 2600-2800 m in the south) has open stands
of P. jeffreyi as the prominent species. The lower limits of
the coniferous forest merge into the Artemisia tridentata
(sagebrush) association of the Basin and Range Province.
Pacific Northwest Region This region consists of the Pa-
cific coast coniferous forest: the Coast Ranges of midcoastal
California; the Cascade Mountains of coastal Oregon, Wash-
ington, and British Columbia; the Olympic Mountains of
northwest Washington; and the British Columbia Coast
Ranges. These forests form a relatively narrow belt along
the western coast of North America from about central Cal-
ifornia to Alaska. They generally extend less than 150 km
inland and range from sea level to several thousand meters
in elevation. They parallel the coniferous forests of the
Sierra Nevada and are bordered on the north by the boreal
forest and Rocky Mountains coniferous forest and on the
south by woodland and scrubland communities.
14 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 1.7. Coniferous forest formation—
montane coniferous forest association (Sierra Nevada
region; Sequoiadendron giganteum), California.
Photograph courtesy of the National Agricultural Library,
Forest Service Photo Collection.
Relatively mild climates are maintained by proximity to
the Pacific Ocean, and temperatures are mostly above -15°C
even in the northernmost portions. Rainfall ranges from
650 mm along the eastern slopes in the south to 4000 mm
in northwestern Washington, producing temperate rainfor-
est vegetation on the Olympic Peninsula. In the south 75%
of the precipitation falls in the winter, and fog is an impor-
tant source of moisture; precipitation is more uniformly
distributed in the north.
Franklin (1988) notes that one characteristic of the Pa-
cific Northwest coniferous forest is the dominance of nu-
merous coniferous species that are the largest and longest
lived representatives of their genera. These include Pseudo-
tsuga menziesii, Tsuga heterophylla, Thuja plicata, Picea
sitchensis, and Sequoia sempervirens (Fig. 1.8). Where
these forests merge with those of the Sierra Nevada, shared
species include Tsuga mertensiana and Picea glauca. Spe-
cies in the Cascade Mountains and in the northern Rocky
Mountains coniferous forest are Tsuga heterophylla, Thuja
plicata, and Taxus brevifolia (Pacific yew). The latter was
earlier a source of taxol used in the treatment of some can-
cers, and damage to the trees from harvesting the bark
threatened the species and the source of the drug. Taxol
was later obtained from the leaves and twigs of other spe-
cies (T. baccata, Europe; T. wallichiana, Himalayas) and is
now being synthesized (Suffness, 1995).
The distribution of several important Pacific Northwest
coniferous elements is being rapidly modified by lumber-
ing. Less than 5% of the coast redwoods remain uncut, and
only about one-half of these are protected in parks. The
cutting of douglas fir in Washington and Oregon accounts
for over one-third of the annual lumber production in the
United States, making douglas fir the most important com-
mercial timber tree in North America (Vankat, 1979).
Franklin (1988) recognizes six major forest types within
the Pacific Northwest coniferous vegetation. The principal
one is the Pseudotsuga menziesii (serai)-Tsuga hetero-
phylla (climax) forest. It ranges from sea level to about
1000 m elevation and includes associates such as Thuja
plicata, Abies grandis (grand fir), Picea sitchensis, Pinus
monticola, P. lambertiana, Calocedrus decurrens, Abies
amabilis (Pacific silver fir), and Taxus brevifolia. The minor
hardwood element includes Alnus rubra (red alder), Acer
macrophyllum, Prunus emarginata (bitter cherry); Populus
trichocarpa (black cottonwood) and Fraxinus latifolia
(Oregon ash) along waterways; and Arbutus menziesii (Pa-
cific madrone), Castanopsis chrysophylla (golden chinka-
pin), and Quercus garryana (Oregon white oak) on drier
sites. Common shrubs are Gaultheria shallon, Holodiscus
discolor, Berberis nervosa, and Rhododendron macro-
phyllum.
The Picea sitchensis-Tsuga heterophylla forest grows
under distinctly coastal, maritime climates from northern
California to Alaska. Associates include Pinus contorta,
Pseudotsuga menziesii, Thuja plicata, Abies grandis, A.
amabilis, Acer macrophyllum, Alnus rubra (on former log-
ging sites); Sequoia sempervirens in the south; Tsuga mer-
tensiana and Chamaecyparis nootkatensis in the north;

and Pinus contorta in swamps and dunes. The shrub layer
consists of Acer circinatum, Rubus spectabilis, Vaccinium
parvifolium, V. alaskaense, and Menziesia ferruginea.
The Sequoia sempervirens forest (Fig. 1.8) occurs along
a zone only -16 km wide from northern coastal California
to southern Oregon.3 Associated with the coast redwood
are Pseudotsuga menziesii and Lithocarpus densiflorus (tan
oak); Tsuga heterophylla, Umbellularia californica (Cali-
fornia bay), and Alnus rubra (moist sites); Picea sitchensis
(coastal); and Arbutus menziesii, Calocedrus decurrens,
and Pinus attenuata (drier sites). Common shrubs are Gaul-
theria shallon, Rhododendron macrophyllum, and Vac-
cinium ovatum.
In northwestern California and southwestern Oregon
the Klamath Mountains mixed evergreen forest is a mosaic
of vegetation derived from the surrounding areas. The com-
mon trees are Pseudotsuga menziesii, mixed with Lithocar-
pus densiflorus, Quercus chrysolepis, Arbutus menziesii,
and Castanopsis chrysophylla.
Coniferous elements from the Sierra Nevada extend into
the southern Pacific Northwest forests and form the Sier-
ran-type mixed conifer forest (Franklin, 1988). Represented
are Pseudotsuga mensiesii, Pinus lambertiana, P. ponder-
osa, Abies concolor, and Calocedrus decurrens. At higher
altitudes subalpine forests and parklands develop, within
which various phases can be recognized such as the Abies
amabilis-Tsuga heterophylla forest, the Abies magnifica
var. shastensis forest, the Tsuga mertensiana forest, and the
subalpine parkland.
As noted earlier, the montane coniferous forest associa-
tions are a comparatively uniform vegetation characteristic
of mid- to high elevations in North America. Similarities in
Setting the Goal: Modern Vegetation of North America 15
Figure 1.8. Coniferous forest
formation—montane coniferous
forest association (Pacific North-
west region; Sequoia semper-
virens forest), Del Norte County,
California. Photograph courtesy of
the National Agricultural Library,
Forest Service Photo Collection.
composition are evident between the Appalachian region
and the western mountains and especially within the west-
ern cordilleras. Associations are recognized, in part, on the
basis of the geographic separation of the several mountain
systems and on altitudinal zonation within individual sys-
tems. Where physiography allows a particular climate to
extend across topographic barriers and connect otherwise
isolated regions, coniferous forest elements in these dis-
junct areas are often shared. For example, Pacific air can
penetrate the coastal sierras at places along the Cascade
Mountains and bring heavy rains and cold temperatures to
the western slopes of the Rocky Mountains near the Co-
lumbia Plateau and adjacent Canada. At such places the
Pacific coastal and Rocky Mountains coniferous forests
share species that elsewhere are used to distinguish the as-
sociations. Cascadian elements in this region of the Rocky
Mountains include Tsuga heterophylla, Thuja plicata, Abies
grandis, Taxus brevifolia, Tsuga mertensiana, Larix lyallii,
Menziesia ferruginea, Oplopanax horridum, Philadelphus
lewisii, Rhododendron albiflorum, Sorbus sitchensis, Vac-
cinium globulare, V. membranaceum, and Xerophyllum
tenax (Peet, 1988). The relationship between the Cascade
and Rocky Mountains coniferous forests is one example of
the variable composition of associations. Clearly any change
in climate or topography would result in greater opportu-
nity for intermingling of the dominant and associated
species between regions currently separated geographi-
cally, and there is no guarantee that after reshuffling the
principal tree'component and associated vegetation would
remain the same. In fact, results from Quaternary pollen
analysis indicate it is unlikely. This raises the question of
how constant paleocommunities remain in composition
16 Late Cretaceous and Cenozoic History of North American Vegetation
through time, which has implications for scenarios at-
tempting to trace the history of major vegetation units. The
fluid composition of various closely related coniferous for-
est associations seem to necessitate an equally flexible
component to any paleoecological concept in order to ex-
plain the history of the North American vegetation. These
ideas will be considered in Chapter 3, but similarities be-
tween the Cascade and Rocky Mountains coniferous
forests in parts of their range provide an introduction to the
temporal and dynamic nature of plant associations.
DECIDUOUS FOREST FORMATION
The eastern deciduous forest formation covers -11% of the
continent, and it is the most diverse and species-rich com-
ponent of the North American vegetation. Whereas the bo-
real forest is characterized by about nine principal tree
genera, at least 67 canopy and subcanopy trees and shrubs
combine in various ways to constitute the associations of
the deciduous forest. The canopy averages ~30 m in height,
although Greller (1988) quotes old accounts of dense virgin
forests in Ohio with oaks, hickories, pines, and other trees
considerably larger than in any modern second-growth
forests. The formation is bounded on the west by grass-
lands from western Minnesota to eastern Texas where
lower moisture and, formerly, fires by Amerindians limited
its distribution. To the north it is bordered by the boreal
forest from Minnesota to northern Maine where it becomes
limited primarily by low temperatures, lower light inten-
sity, and a shorter growing season. These and other factors
combine to render deciduous plants at a competitive dis-
advantage to evergreen plants such as conifers. Deciduous
components blend with boreal coniferous species in the
upper midwest to form a transitional vegetation called the
lake states forest. Elements of the deciduous forest are also
found in western North America where they are most com-
mon as riparian vegetation along river valleys and flood-
plains (e.g., Salix, Alnus, Populus}.
Precipitation is ~1250 mm along the Atlantic coast, 850
mm along the forest-grassland boundary, 750 mm in the
Great Lakes region, and 1500 mm along the Gulf Coast. Av-
erage January temperatures range from -6.7°C in the north
to 4.4°C in the south; average temperatures in July are 21.1
and 26.7°C, respectively. Under these temperature regimes
the growing season is 120 days in the north and 250 days
in the south (Vankat, 1979).
Several systems are available for classifying the associa-
tions of the deciduous forest formation (Braun, 1950;
Greller, 1988; Vankat, 1979). Following Vankat (1979), eight
associations are recognized. The mixed mesophytic associ-
ation occurs on moist, well-drained sites in the central
parts of the formation in the unglaciated Appalachian
Plateau and Cumberland Mountains of eastern Tennessee
and Kentucky. This is the richest and most diverse associa-
tion and includes as dominants Fagus grandifolia (Ameri-
can beech), Liriodendron tulipifera (tulip tree), Tilia (bass-
wood, several species), Acer saccharum (sugar maple), Aes-
culus octandra (buckeye), Quercus rubra (red oak), Q. alba
(white oak), Tsuga canadensis (eastern hemlock), plus 30 or
more others that commonly form more local associations.
The beech-maple association (Fig. 1.9) is restricted pri-
marily to glaciated areas of Ohio, western Indiana, and
southern Michigan. The dominants are Fagus grandifolia
and Acer saccharum, and species of Fraxinus (ash) and
Ulmus americana (American elm) are common associates.
The maple-basswood association occupies the smallest
area, occurring principally in east-central and southeastern
Minnesota, southwestern Wisconsin, and northeastern Iowa.
The boundaries are not clear and it has been suggested that
this is a consequence of former burning practices by the
Amerindians. The dominants are Acer saccharum and
Tilia americana (replacing Fagus grandifolia}, and various
species of oak are subdominants.
The lake states forest association is a hemlock-white
pine-northern hardwood association and, as noted previ-
ously, it is transitional between the deciduous forest and
the boreal coniferous forest. It is distributed from western
Ontario and Minnesota, across southern Canada and New
England, to the Maritime Provinces of eastern Canada. At
different places the association may be dominated by conif-
erous species, deciduous species, or a combination of both.
Widespread members include Pinus strobus (white pine), P.
resinosa (red pine), and Betula lutea. Species with boreal
affinities are Betula papyrifera, Populus tremuloides, and
Pinus banksiana. Deciduous forest elements are Acer
rubrum (red maple), A. saccharum, Fagus americana, Tilia
americana, and Tsuga canadensis. In stands dominated by
conifers the principal species are Pinus strobus, P. resinosa,
and P. banksiana on drier sites; the boreal species Picea
glauca, Picea mariana, and Larixlaricina are in and around
scattered bogs. To the east Picea rubens and Tsuga sp. are
common. Stands of white pine were formerly much more
extensive, but they have been intensively lumbered since
about 1623, initially to provide masts for British sailing
ships, and have been replaced mostly by hardwoods.
The oak-chestnut association no longer exists in its
original form because of the virtual elimination of mature
trees of Castanea dentata (American chestnut) by the as-
comycete fungus Endothia parasitica. The parasite was in-
troduced from China in about 1900 and was first reported
on American trees in New York City in 1904 (Greller, 1988;
Vankat, 1979). By 1920, 50% of the chestnut trees in the
north had been destroyed, and by 1930 50% were gone
from southern forests. Castanea dentata is now primarily a
scattered understory tree reproducing by vegetative meth-
ods, and it has been replaced mostly by oaks (Quercus bo-
realis var. maxima, Q. prinis, and Q. alba).
The most widespread association in the eastern decidu-
ous forest formation is the oak-hickory association, ex-
tending from New Jersey south along the piedmont to the
Gulf states, west to central Texas, and north to Minnesota.

At its western edge it grades into the grassland formation
as a savanna. The dominant oaks in the south and south-
west are Quercus stellata (post oak) and Q. marilandica
(blackjack oak); in the west are Quercus alba (white oak),
Q. rubra, Q. velutina, and Q. macrocarpa (bur oak). The
hickories are Carya cordiformis (bitternut hickory), C. to-
mentosa (mockernut hickory), C. ovalis (red hickory), and
C. ovata (shagbark hickory). The region of greatest species
diversity is in the low hills of the Ozark Plateau and Oua-
chita Mountains of central Arkansas. In the east and south
and the piedmont region of Georgia, pines become an im-
portant component. The settlement of Georgia's rich pied-
mont began in 1773, and by 1840, 87% of the area was
under cultivation. During the Civil War -10% of the land
was abandoned, followed by another 30% during the agri-
cultural depression of the late 1880s, and 35% following
spread of the boll weevil in the 1920s (Vankat, 1979). Most
of the abandoned land was regenerated by pine. In the
south the prominent species are Pinus taeda (loblolly pine)
and P. echinata, while in the north P. rigida (pitch pine)
and P. virginiana (Virginia pine) are the dominants. The
pine phase is fire dependent, and without fire hardwoods
replace them in most areas of the oak-hickory association.
The southern mixed hardwood association is scattered
Setting the Goal: Modern Vegetation of North America 17
Figure 1.9. Deciduous forest formation—
beech-maple association (Fagus grandifolia, Acer
saccharum), Nicolet National Forest, Wisconsin.
Photograph courtesy of the National Agricultural
Library, Forest Service Photo Collection.
along the coastal plain from North Carolina to Texas, and
various phases are influenced by the level of the water table
and the water-retention capacity of the soil. It is a mul-
tidominant assemblage and individual stands may have
from five to nine prominent tree species. The more common
ones in well-drained areas with rich soils are Fagus ameri-
cana, Quercus alba, Q. virginiana (live oak), Q. laurifolia
(laurel oak), and Magnolia grandiflom (evergreen magnolia).
Pine woods are locally developed on deep sandy soils
of the coastal plain and include the northern pine barrens
that extend south to near the Delaware Bay (Christensen,
1988). Pinus rigida is the dominant, mixed with P. echi-
nata, Comptonia peregrina, Kalmia latifolia, Gaylussacia
baccata, G. frondosa, G. dumosa, Quercus ilicifolia, Q. pri-
noides, Ilex glabra, and Clethra alnifolia. To the south is
the sandhill pine forest of Pinus palustris (longleaf pine)
and the understory grass Aristida stricta (wire grass). Asso-
ciated vegetation defining three phases of the forest in-
cludes Quercus laevis (turkey oak), Nyssa sylvatica (black
gum), Diospyros virginiana (persimmon), Lyonia mariana
(staggerbush), and Gaylussacia dumosa (dwarf huckle-
berry)—the pine-turkey oak sandridge phase; Quercus
marilandica, Q. margaretta (sandhill post oak), Q. indica
(bluejack oak), and Liquidambar styraciflua (sweet gum)—
18 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 1.10. Deciduous forest
formation—flood-plain forest
association (Taxodium distichum],
Louisiana. Photograph courtesy
of the National Agricultural
Library, Forest Service
Photo Collection.
the fall-line sandhill phase; and Q. laevis, Q. marilandica,
Q. indica, Q. rubra, Pinus elliottii (slash pine), and Diospy-
ros virginiana—the Florida sandhill phase.
Another pineland along the southeastern coastal plain
is the sand pine scrub of Pinus clausa (sand pine) and as-
sociated vegetation of Serenoa repens (saw palmetto), S.
etonia (scrub palmetto), Quercus germinata, Q. myrtifolia,
Q. virginiana, Q. inopina, Gary a floridana, Lyonia ferrug-
inea, and Ceratonia ericoides.
The flood-plain forest association (Fig. 1.10) is a riparian
vegetation. A southern phase is characterized by Taxodium
distichum (bald cypress), Chamaecyparis thyoides (south-
ern white cedar), Pinus glabra (spruce pine), Nyssa (tupe-
los, N. sylvatica, N. aquatica, N. ogeche), Quercus lyrata
(overcup oak), Q. laurifolia, Q. michauxii (basket oak), Q.
falcata (cherry bark oak), Q. phellos (willow oak), Carya
aquatica (water hickory), Ilex verticillata (winterberry),
I. opaca (American holly), Gleditsia aquatica (water lo-
cust), Asimina triloba (pawpaw), Lindera benzoin (spice-
bush), Sabal palmetto (palmetto), Itea virginica, Styrax
americanum, Cornus foemina, Acerrubrum, Ulmus amer-
icana, Fraxinus caroliniana, F. profunda, F. pennsylvan-
ica, Magnolia virginiana (sweet bay), Persea borbonia (red
bay), Cyrilla racemiflora, Leucothoe racemosa, Lyonia lu-
cida, Clethra alnifolia, and Liquidambar styraciflua (sweet
gum). In the north, floodplain habitats are occupied by
species ofPopulus, Salix, Ulmus, and Fraxinus.
The deciduous forest formation is found in several
disjunct regions throughout the northern hemisphere, in
addition to outliers in the American west. The forest is es-
pecially well developed in eastern Asia, and a phase inter-
mixed with warm temperate to tropical elements occurs
along the eastern escarpment of the Mexican Plateau. It was
also extensive in central and western Europe during the
Tertiary, disappearing there as a result of climatic changes
in the Late Tertiary and the continental glaciations of the
Pleistocene. One of the goals of Cenozoic vegetational his-
tory studies is to better understand the origin of such dis-
junct floristic relationships (Chapter 9); these attempts are
facilitated by the fact that almost all of the tree genera,
many of the shrubs, and some of the herbs mentioned in the
above discussion are represented in the fossil record.
GRASSLAND FORMATION
When French explorers moved south from eastern Canada
into the central plains, they encountered a vast expanse of
grassland never before witnessed by western Europeans,
and applied the closest French word for a community dom-

inated by grasses and forbs—prairie (meadow). The grass-
land is the most extensive of the North American plant for-
mations, originally covering an estimated 265,987,700 ha of
the 770 million ha in the United States (-30%, Sims, 1988;
-21% of North America north of Mexico, Barbour and
Christensen, 1993). It extends from Indiana south along the
western edge of the deciduous forest to the Front Ranges of
the Rocky Mountains and from southern Canada through
northern Mexico. The grassland exists, in part, as a result of
insufficient precipitation during the spring and summer
growing season to support trees. Precipitation decreases
from east to west and from north to south. Annual rainfall is
-1000 mm in the tallgrass prairie of the northeast and -200
mm in the desert grasslands of the southwest. Annual tem-
peratures range from 3°C in mountain grasslands to 15°C in
the desert grasslands. One of the characteristics of the North
American grassland climate is the distinctly seasonal rain-
fall and the extremes in temperature. In fact, the range of
climatic parameters in the grassland is greater than for any
other North American plant formation (Sims, 1988). In the
central part of the continent temperatures may be -35°C in
the winter and reach 45°C in the summer. The growing sea-
son is from 120 days in the north to 200 days in the south
and extends virtually year-round in the southwestern
United States and northern Mexico. In the California grass-
lands rainfall is 150-500 mm and falls mostly during the
winter months. Fire is an important, if not essential, factor
in the maintenance of the modern North American grass-
land (Collins and Wallace, 1990; Wright and Bailey, 1982).
The front of one fire early in this century covered over 200
km in a single day (Vankat, 1979).
The grasslands include -600 genera and 7500 species of
plants (Sims, 1988). About 95% are perennial, and the
Leguminosae-Fabaceae and Compositae-Asteraceae con-
tribute the largest number of associated forbs. Another
characteristic of the grassland formation is the dramatic
change in composition that often takes place throughout
the growing season. In one prairie in Wisconsin, 17 new
species came into flower each week from April through Oc-
tober (Vankat, 1979).
Six associations are recognized within the grassland for-
mation: the tallgrass prairie, mixed grass prairie, and short-
grass prairie in the central part of the continent; the desert
grassland of the arid southwest in Arizona, New Mexico,
Texas, and northern Mexico; the Palouse Prairie in parts of
Washington, Oregon, and Idaho; and the California grass-
land in the Central Valley of California between the Coast
Ranges and Sierra Nevada.
Grasses of the tallgrass prairie association may reach 3 m
in height near the end of the growing season. This is the most
fertile of the associations and, because of its agricultural
value for growing wheat and corn, little of the native tallgrass
prairie remains today (Samson and Knopf, 1994). The tall-
grass prairie is the most mesic of the grassland associations
and is best developed along the margin of the deciduous for-
est. A lobe extending eastward into Indiana and Illinois is
Setting the Goal: Modern Vegetation of North America 19
known as the Prairie Peninsula (Stuckey and Reese, 1981;
Transeau, 1935). The principal grasses are Andropogon ger-
ardi (big bluestem), Sorghastrum nutans (Indian grass), Pan-
icum virgatum (switchgrass), Elymus canadensis (Canada
dry grass), and Spartina pectinata (prairie cordgrass, mainly
in wet, poorly drained areas). In the drier uplands are An-
dropogon scoparius (little bluestem), Stipa spartea (needle-
grass), Sporobolus heterolepis (prairie dropseed), Bouteloua
curtipendula (sideoats grama), and Koeleria cristata (June-
grass). An elongated isolated lobe called the blackland
prairie (Andropogon, Stipa) runs from north of Dallas, to
south near Austin, to San Antonio (Kuchler, 1964).
The mixed grass association (Fig. 1.11) lies between the
tallgrass and shortgrass association. The short phase of the
association includes Bouteloua gracilis (blue grama), B.
hirsuta (hairy grama), and Buchloe dactyloides (buffalo
grass). The taller phase consists of Andropogon scoparius
and Stipa comata (needle and thread grass).
The westernmost portion of the central grasslands, a
physiographic area known as the High Plains, is occupied
mostly by the shortgrass association that extends into Mex-
ico as far as Jalisco (22° N). Rainfall is less than in the tall-
grass and mixed grass associations and the plants typically
reach only 20-50 cm in height. The dominants are Boute-
loua gracilis and Buchloe dactyloides, associated with
Stipa comata, Aristida longiseta (wire grass), Agropyron
smithii (western wheat grass), Sporobolus cryptandus,
Muhlenbergia torreyi, Koeleria cristata, and Hillaria jame-
sii. Before the advent of more rational farming practices,
wind erosion of the soil was extensive and created the
great dust bowl storms of the 1930s. In March 1935, 37,000
metric tons of soil per cubic kilometer of air was recorded
above Wichita, Kansas, that came from an area over 400 km
away. In some areas vegetation cover during this period
was reduced by 98% (Vankat, 1979).
The desert grassland association occupies plateaus in
the arid southwest region of west Texas, southern New
Mexico, Arizona between the desert and pinon-juniper
woodlands, and into Mexico (McClaran and Van Devender,
1995). The prominent grasses are Bouteloua eriopoda (grama
grass), Hilaria belangeri (tobasa grasses), Oryzopsis hymen-
oides, Muhlenbergia porteri, and Aristida sp. However, dry-
ing climates and recent overgrazing have caused shrubs to
invade from adjacent desert vegetation, and much of the
area is occupied by Prosopis glandulosa (mesquite), Larrea
tridentata (creosote bush), Acacia (acacias), Yucca (yucca,
Spanish bayonet), and Opuntia (prickly pear).
The California grassland association in the agriculturally
rich Central Valley now includes mostly introduced annual
grasses and other plants; between 50 and 90% of the present
vegetation cover in uncultivated areas is comprised of exotic
species. Cool-season perennials have been replaced by cool-
season annuals, partially because of the grazing of perenni-
als in preference to the lower quality forage of many of the
annuals. The original dominant was Stipa pulchra (needle-
grass), which has now been replaced by such grasses as
20 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 1.11. Grassland formation—mixed grass association (Bouteloua gracilis), Colorado. Photograph courtesy of the
National Agricultural Library, Forest Service Photo Collection.

Avenafatua (weedoats), Hordeum murinum (mouse barley),
Bromus sp. [brome grasses), and Festuca sp. (fescues).
The Palouse Prairie association developed as a result of
the rainshadow created by uplift of the Cascade Moun-
tains. Originally the dominants were Agropyron spicatum
(bluebunch wheat grass), Festuca idahoensis (Idaho fes-
cue), Elymus condensatus, and Poa secunda (sandburg
bluegrass). In the late 1800s the annual Bromus tectorum
(downey chess) was introduced and it is now the domi-
nant, along with Artemisia tridentata (sagebrush) that has
come in with human disturbance.
SHRUBLAND/CHAPARRAL-WOODLAND-
SAVANNA FORMATION
Mediterranean and Madrean Scrublands
and Woodlands
The vegetation characteristic of lower elevations and drier
parts of the western intermountain basins, valleys, slopes,
and plateaus consists of three phases based on the spacing
or density of the tree or shrub layer. Shrubland/chaparral
or thickets are composed of a dense, mostly continuous
layer of shrubs and small shrubby trees. Woodlands (Fig.
1.12) are vegetation composed principally of trees in
which the crowns do not touch. Savanna is a vegetation in
which the tree or shrub cover is typically less than 30%
and a dense herbaceous layer usually dominated by grasses.
All of these developed primarily in response to drier con-
ditions created to the lee of the rising Sierra Nevada and
Cascade Mountains. They cover -1% of North America
north of Mexico.
The scrub oak-mountain mahogany shrubland occurs
from southern Wyoming and northern Utah to Mexico at
elevations between 1250 and 1750 m and is typically
found as a band between the coniferous forest and desert
vegetation. The composition of the association varies con-
siderably. The most wide-ranging species is Cercocarpus
ledifolius (curl-leaf mountain mahogany). The greatest di-
versity is found in the vicinity of the southern Rocky
Mountains where the dominants are Quercus turbinella

Figure 1.12. Shrubland/chaparral-woodland-savanna formation—juniper (Juniperus occidentalis) woodland, Cen-
tral Oregon. Photograph courtesy of The National Archives.
(scrub oak), Q. emoryi, Q. dumosa, and Q. chrysolepis (an
eastern extension of a California oak group south of the
Colorado Plateau). In the north the species are Quercus
gambellii (Gambell's oak), Q. undulata (a hybrid between
Q. gambelii and several southern evergreen live oaks;
Tucker, 1961), Cercocarpus breviflorus, C. betuloides, and
C. montanus. Other shrubs include Artemisia tridentata
var. vaseyana, Acer grandidentatum, Purshia tridentata,
Rhus triloba, Rhamnus corcea, Fallugia paradoxa, Cowa-
nia mexicana, Arctostaphylos pungens, A. pringlei, Cean-
othus greggii, Garrya flavescens, G. wrightii, Eriodictyon
angustifolium, and species of Amelanchier, Symphoricar-
pus, and Berberis (West, 1988).
Chaparral is a version of the shrubland found mostly in
California from southern Oregon (Rogue River watershed,
43° N latitude) to northern Baja California (Sierra San
Pedro Martir, 30° N latitude; Keeley and Keeley, 1988).
Chaparral and related shrub communities also extend
along the Sierra Madre Occidental to southern Durango,
Mexico, as summer-wet Madrean vegetation. It is a dense,
sclerophyllous vegetation most common in California be-
tween the foothills of the Sierra Nevada and the Pacific
coast from sea level to -2000 m elevation. The shrub layer
is 1-4 m tall and is frequently so dense that it is virtually
impenetrable. The word chaps is derived from chaparral,
Setting the Goal: Modern Vegetation of North America 21
denoting pieces of leather tied across the legs for protec-
tion while riding through this vegetation. The area is char-
acterized by the cool wet winters and hot dry summers
typical of a Mediterranean climate (Arroyo et al., 1995).
Precipitation ranges from 200 to 1000 mm, and 80-90%
falls from November to April. The mean winter minimum
temperature is between 0 and 10°C, while the summer
maximum reaches 40°C. If summer fires are frequent, the
vegetation trends toward grassland. If they are less fre-
quent, the development of oak woodland is favored
(Vankat, 1979).
The composition of the California chaparral is variable
from one locality to another and may include from one to
as many as 20 dominants. Keeley and Keeley (1988) pre-
sent a table listing 63 of the common species. The most
widely distributed member is Adenostoma fasiculatum
(chamise), frequently associated with Salvia mellifera, S.
alpina (California coastal sage), Eriogonum fasciculatum,
and species of Arctostaphylos (manzanitas) and Ceono-
thus. The range of the California chaparral is expanding
as a result of overgrazing, mining, burning, logging, and
construction.
The broad-sclerophyll forest has a range that mostly cor-
responds to that of the California chaparral. It occurs on
more mesic sites and has a tree layer that is from 5 to 10 m
22 Late Cretaceous and Cenozoic History of North American Vegetation
in height compared to 1-4 m for the chaparral. It is also
more common than chaparral in the north where annual
rainfall is greater (1000 mm) and less common in the south
where rainfall within its range is -350 mm. Mean annual
temperatures are below freezing in the winter to over 30°C
in the summer. The broad-sclerophyll forest extends to
-1500 m elevation where it is commonly bordered by the
lower montane coniferous forest. The most prominent
species are Quercus dumosa (scrub oak), Q. chrysolepis
(canyon live oak), Q. agrifolia (coast live oak), Q. wislizenii
(interior live oak), Arbutus menziesii (madrone), Pinus
sabiniana (digger pine), P. coulteri (Coulter pine), Umbel-
lularia californica, Aesculus californica (California buck-
eye), Heteromeles arbutifolia (chaparral holly), Prunus
lilcifoia (chaparral cherry), Cercocarpus betuloides, and
Malosma (Rhus) laurina (laurel sumac).
The California woodland vegetation is found in the in-
terior valleys of California from near Los Angeles to north-
ern California at elevations between 150 and 1000 m. The
forest consists of trees from 5 to 10 m in height with an
open canopy covering 30-50% of the ground. The domi-
nants are Quercus douglasii (blue oak), Q. lobata (Califor-
nia white oak), and Q. agrifolia (coast live oak).
The pifion-juniper woodland is best developed in the
Great Basin and in the drier parts of the southern Rocky
Mountains. It extends to the eastern slope of the Sierra
Nevada and from Arizona and New Mexico through south-
ern and eastern Colorado, Utah, and Nevada, to southern
Idaho (Miller and Wigand, 1994; West, 1988). The associa-
tion has a discontinuous distribution along the high mesas,
foothills, and lower elevations of large mountain systems
and a continuous range over many of the smaller moun-
tains. In most areas the pifion-juniper woodland occurs at
elevations between 1500 and 2250 m. Upslope it is fre-
quently bordered by ponderosa and other pine forests and
at lower elevations by Artemisia shrubland, desert, or grass-
land. It is an extensive association covering -24 million ha
in the western United States and ~19 million ha in the in-
termountain region between the Sierra Nevada-Cascade
Mountains and the Rocky Mountains. Rainfall is between
250 and 500 mm, and mean annual summer temperatures
are ~20°C with highs to 30°C. The structure and composi-
tion of this woodland is comparatively simple and in-
cludes as the principal dominants Pinus edulis, P. mono-
phylla, P. cembroides, Juniperus erythrocarpa (redberry
juniper), /. occidentalis (western juniper), /. scopulorum
(Rocky Mountains juniper), /. monosperma, J. deppeana
(alligator juniper), and/, osteosperma (Utah juniper).
The savanna in North America is a fire-dependent, pri-
marily grassland-deciduous forest boundary vegetation. It
also includes open pinelands in the southeastern United
States, areas of the central valley of Oregon, and the more
open parts of the California woodland vegetation. Before
human disturbance, in the east it extended as a 75-175 km
wide belt from Minnesota to Texas. In composition it is a
mixture of grassland and oak-hickory association species,
including Quercus macrocarpa in the north and Q. stellata
and Q. marilandica in the south. Further south in Texas it
borders the blackland prairie, and Juniperus and Prosopis
are common associates (Kiichler, 1964). Fire is important to
the perpetuation of the savanna; because fire is now con-
trolled in many areas, woodland or forest vegetation has
invaded areas once occupied by oak savanna.
DESERT FORMATION
Deserts are areas receiving less than -120 mm of available
precipitation per year. Most North American deserts re-
ceive slightly more and are only semiarid. The annual rain-
fall is characteristically variable in amounts, and it is
unevenly distributed throughout the year. Temperature re-
gimes differ among the different kinds of deserts (warm
deserts versus cold deserts), but at least some season of the
year has high temperatures. Another feature of the desert is
high winds that, combined with low and unpredictable
rainfall, high temperatures, and high light intensities, pro-
duce the most biologically significant facet of desert cli-
mates: a low precipitation/evaporation ratio. Even during
the rainy period there is usually rapid runoff or rapid per-
colation of water through the coarse soils, particularly in
summer rain deserts (e.g., Chihuahuan). These areas re-
ceive rain in the form of brief and intense thunderstorms,
in contrast to winter rain deserts (e.g., Mojave) where pre-
cipitation periods are longer and less intense (MacMahon,
1988). As a consequence, there has been evolutionary se-
lection and elaboration of characters and processes that
compensate for water stress, and these constitute the syn-
drome of features typically associated with desert plants:
low stature, wide spacing, alleopathy, deep penetrating or
extensive shallow root systems, coriaceous and microphyl-
lous leaves, pubescense, replacement of leaves by spines,
presence of leaves only during the rainy season, succu-
lence, pleating of stems to accommodate rapid changes in
volume with sudden water availability, slow growth rate,
water-soluble growth-inhibiting substances in the seed
coat, rapid reproductive cycle of annuals, and others. With
these adaptations many desert plants withstand ambient
temperatures in excess of 55°C where, for example, pubes-
cence can reduce plant surface temperatures by 13°C
(Vankat, 1979). The effectiveness of these modifications is
apparent from an observation cited in MacMahon (1988)
that water loss from a plot of bare soil and one containing
Larrea tridentata was virtually the same.
There are four deserts on the North American continent
(-5% of North America north of Mexico; Fig. 1.13, Table 1.4),
in addition to the polar desert of the Arctic. These are the
Great Basin, Mojave, Sonoran, and Chihuahuan deserts. The
Great Basin is a cold desert receiving -60% of its precipita-
tion as snow, and it is often included in the shrubland vege-
tation of the Basin and Range Province. The others are warm
deserts in the southwestern United States and adjacent Mex-
 Setting the Goal: Modern Vegetation of North America 23

Figure 1.13. Distribution of North American deserts.

Reprinted from MacMahon (1988) with the permission of
Cambridge University Press.

ico, receiving most of their precipitation in the form of sum-
mer or winter rains (MacMahon and Wagner, 1985).
The Great Basin Desert association corresponds in dis-
tribution to the Great Basin physiographic province, occur-
ring mostly in the zone between the Sierra Nevada-
Cascade Mountains and the Rocky Mountains at elevations
between 1200 and 2500 m (Osmond and Pitelka, 1990). It
is a result of the rainshadow created by the Sierra Nevada
and Cascade Mountains. The Great Basin Desert is also typ-
ical of broad valley floors that receive precipitation from
runoff that slowly evaporates, producing salt-rich playa
lakes. Annual rainfall is between 100 and 250 mm. Winter
temperatures usually do not go above freezing, and sum-
mer frosts occur at night in the valleys, but during the sum-
mer day the temperatures may reach above 40°C. At higher
elevations the association grades into pifion—juniper wood-
land. Cacti and other succulents typical of the warm des-
erts are less common in the Great Basin Desert.
Areas of low soil salinity support stands of Artemisia
tridentata, the most widespread vegetation type in the in-
termountain lowlands. In soils of high salinity Atriplex
confertifolia (shadscale) is the conspicuous element. Com-
mon associates are Artemisia spinescens, Atriplex nuttallii
(Nuttall saltbush), Atriplex spinosa, Salicornia utahensis,
Suaeda fruticosa, S. torreyana, Sarcobatus vermiculatus
(greasewood), Eurotia lantana (winter fat), and several
perennial grasses such as Festuca idahoensis, Agropyron
(Pseudoroegneria) spicatum, A. (Pascopyrum) smithii, Ely-
mus lanceolatus, Stipa thungergia, S. arida, and S. spe-
ciosa. Overgrazing has increased the area covered by sage-
brush, and the annual grass Bromus tectorum has become
widespread along with Chrysothamnus (rabbit bush),
Ephedra (Morman tea), and Tetradymia (horse brush). The
term steppe or woody steppe is primarily a European term
for temperate shrub vegetation with plentiful short grasses
often growing on dry seasonally cold plains and mountain
slopes. One difference between savanna-grassland and
steppe is that the latter is summer dry.

Table 1.4. Approximate areas of North American deserts.

Unit Area (km2) Portion of NA Desert Area (%)

Great Basin Desert 409,000 32.0

Sonoran Desert 275,000 21.5
Mojave Desert 140,000 11.0
Chihuahuan Desert 453,000 35.5
Total desert 1,277,000 100
Warm deserts 868,000 68.0
Cold deserts 409,000 32.0
Basin and range province 1,717,000 —

Adapted from MacMahon (1988).

24 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 1.14. Desert Formation—Sonoran Desert association
(Carnegiea gigantea, Larrea tridentata). •&;
The Mojave Desert association is the smallest of the
North American deserts, extending from southern Califor-
nia through southern Nevada to western Arizona. Eleva-
tion is from -86 m in Death Valley to 1200 m, and precipi-
tation ranges from 125 mm in the western portions to 50
mm in the east. Winter temperatures reach 15°C during the
day and are below freezing at night, while summer temper-
atures are from 30° to 40°C in the basins to 18° to 30°C on
mountain summits.
The Mojave Desert is often considered a transition be-
tween the Great Basin Desert to the north and the Sonoran
Desert to the south. However, MacMahon (1988) notes that
-25% of the total species and 80% of the -250 annuals are
endemic. The dominants are Larrea tridentata (creosote
bush) and Franseria (Ambrosia) dumosa (bur sage), which
cover -75% of the region. In the north Artemisia tridentata
and Atriplex confertifolia come in at the higher elevations,
and in the west Yucca schidigera (Mojave yucca) and Y.
brevifolia (Joshua tree) are characteristic members of the
vegetation. In the Mojave Desert, like the Great Basin Des-
ert, cacti are present but are not as common as in the Chi-
huahuan Desert.
The Sonoran Desert association (Fig. 1.14) is the most
varied of the North American deserts in terms of tempera-
ture, rainfall, and composition. It extends from southern
California and southern Arizona into Baja California and
northwestern mainland Mexico. Most of the area consists of
plains and moderate mountainside slopes called bajadas,
which formed by the coalescence of alluvial fans. It is the
lowest of the North American deserts, occupying sites gen-
erally between sea level and 900 m, but mostly below 600
m. The Sonoran Desert is not a rainshadow desert, but it is
maintained primarily by a persistent high-pressure system.
Precipitation is variable and irregular. Rainfall ranges from
near 0 mm in the west to 35 mm in the east, to as much as
700 mm on some low mountain peaks. In the summer tem-
peratures reach a maximum of 40°C, winter highs are
around 20°C, and frosts are rare. This makes the Sonoran
Desert the hottest and most temperature variable of the
North American deserts; on bare rock daily temperatures
may fluctuate by 60°C within a 24-h period.
These variations in climate also make this desert the
most diverse in composition (Shreve and Wiggins, 1964).
The region around Tucson, Arizona has many small trees
and shrubs related to those of the Sinaloan (Mexico) thorn
forest. In the arid lowlands Franseria (Ambrosia) dumosa
and Larrea tridentata are the most common species, but
each has a varied associated vegetation. These include
Ephedra trifurca, several species of Dalea, Sapium bilocu-
lare, Condalia lycioides, Prosopis glandulosa, and Encelia
farinosa (brittlebush). At higher elevations Cerdidium mi-
crophyllum (palo verde), Fouquieria splendens (ocotillo),
Acacia greggii, and Carnegiea gigantea (saguaro, Fig. 1.14)
are conspicuous elements.
The Chihuahuan Desert association occurs from south-
ern New Mexico and western Texas into north-central Mex-
ico. In elevation it extends from -400 m along the Rio
Grande to 1500 m on the central Mexican Plateau. The
landscape consists mostly of a plain dissected by valleys
and numerous small sierras. Precipitation (150-400 mm)
is slightly higher than in the other deserts, and most rain
falls in the summer between June and September. Summer
temperatures average 5-10°C lower than in the Sonoran
Desert; winter temperatures are also lower, and night frosts

Figure 1.15. Warm-temperate to subtropical vegetation, Everglades National Park, Florida. Photograph courtesy of
Sammantha Ruiz, National Park Service, Everglades National Park.
are more common. The dominants are Flourensia cernua
(tarbush), Acacia schaffneri, Mimosa biuncifera, Larrea tri-
dentata, Fouquieria splendens, Agave lechuguilla (lechu-
guilla), Koeberlinia spinosa (allthorn), Hechtia scariosa,
Leucophyllum frutescens, Prosopis sp., Yucca filifera, Y.
carnerosana, and Agave salmiana. Cacti and succulents
are abundant and form the distinctive element of the Chi-
huahuan Desert. They include Myrtillocactus geometri-
zans, Opuntia robusta, O. leucotricha, O. streptacantha,
O. cantabrigiensis, O. phaeacantha, O. leptocaulis, and
Echinocactus horizonthalonius.
TROPICAL ELEMENTS
A tropical formation does not occur in the region covered
in this survey. Rather, elements of tropical vegetation ex-
tend various distances into peninsular Florida and along
the southern Atlantic and Gulf Coast. The most distinctive
are the mangroves (black, Avicennia nitida; white, Lagun-
cularia racemosa; and red, Rhizophora mangle) often asso-
ciated with Conocarpus erecta (buttonwood; Myers and
Ewel, 1990; Odum and Mclvor, 1990). With approximately
80% of the Florida coastline commercially developed, the
Setting the Goal: Modern Vegetation of North America 25
most extensive remnants of this community are found in
the protected habitats of the Everglades National Park (Fig.
1.15) and the J. N. "Ding" Darling National Wildlife Refuge
on Sanibel Island (Fig. 1.16).
SUMMARY
These seven formations constitute the modern vegetation
of North America. They are the product of environmental-
biological interactions over mostly the past 70 m.y, and in
composition and distribution they represent only the latest
of many previous and future versions. To better understand
this vegetation, it is necessary to know some of the factors
that have shaped its history.
References
Arroyo, M. T. K., P. H. Zedler, and M. D. Fox (eds.). 1995.
Ecology and Biogeography of Mediterranean Ecosys-
tems in Chile, California and Australia. Springer-
Verlag, Berlin.
Axelrod, D. I. and P. H. Raven. 1985. Origins of the
cordilleran flora. J. Biogeog. 12: 21-47.
26 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 1.16. Tropical elements—J. N. "Ding" Darling National Wildlife Refuge, Sanibel Island, FL (red mangrove,
Rhizophora mangle). Photograph courtesy of Kathleen Blase.
Barbour, M. G. 1988. Californian upland forests and wood-
lands. In: M. G. Barbour and W. D. Billings (eds.),
North American Terrestrial Vegetation. Cambridge
University Press, Cambridge, U.K. pp. 131-164.
and W. D. Billings (eds.). 1988. North American Ter-
restrial Vegetation. Cambridge University Press, Cam-
bridge, U.K.
and N. L. Christensen. 1993. Vegetation of North
America north of Mexico. In: Flora of North America
Editorial Committee (eds.), Flora of North America
North of Mexico. Oxford University Press, Oxford,
U.K. Vol. 1, pp. 97-131.
Bliss, L. C. 1988. Arctic tundra and polar desert biome. In:
M. G. Barbour and W. D. Billings (eds.), North Ameri-
can Terrestrial Vegetation. Cambridge University
Press, Cambridge, U.K. pp. 1-32.
Braun, E. L. 1950. Deciduous Forests of Eastern North
America. Blakiston Co., Philadelphia, PA.
Christensen, N. L. 1988. Vegetation of the southeastern
coastal plain. In: M. G. Barbour and W. D. Billings
(eds.), North American Terrestrial Vegetation. Cam-
bridge University Press, Cambridge, U.K. pp. 317-363.
Collins, S. L. and L. L. Wallace. 1990. Fire in North Ameri-
can Tallgrass Prairies. University of Oklahoma Press,
Norman, OK.
Crutzen, P. J. and J. G. Goldammer. 1993. Fire in the Envi-
ronment. Wiley, New York.
Elliott-Fisk, D. L. 1988. The boreal forest. In: M. G. Barbour
and W. D. Billings (eds.), North American Terrestrial
Vegetation. Cambridge University Press, Cambridge,
U.K. pp. 33-62.
Franklin, J. E 1988. Pacific Northwest forests. In: M. G. Bar-
bour and W. D. Billings (eds.), North American Terres-
trial Vegetation. Cambridge University Press, Cam-
bridge, U.K. pp. 103-130.
Coward, S. N., C. J. Tucker, and D. G. Dye. 1985. North
American vegetation patterns observed with the
NOAA-7 advanced very high resolution radiometer.
Vegetatio 64: 3-14.
Greller, A. M. 1988. Deciduous forest. In: M. G. Barbour
and W. D. Billings (eds.), North American Terrestrial
Vegetation. Cambridge University Press, Cambridge,
U.K. pp. 287-316.
Johnson, E. A. 1992. Fire and Vegetation Dynamics, Studies
from the North American Boreal Forest. Cambridge
University Press, Cambridge, U.K.
Keeley, J. E. and S. C. Keeley. 1988. Chaparral. In: M. G. Bar-
bour and W. D. Billlings (eds.), North American Terres-
trial Vegetation. Cambridge University Press, Cam-
bridge, U.K. pp. 165-207.
Kiichler, A. W. 1964. Manual to Accompany the Map Po-
tential Natural Vegetation of the Conterminous United
States. American Geographic Society, New York. Spe-
cial Publ. 36.
Lowe, C. H. 1980. Arizona's Natural Environment. Univer-
sity of Arizona Press, Tucson, AZ.

MacMahon, J. A. 1988. Warm deserts. In: M. G. Barbour and
W. D. Billlings (eds.), North American Terrestrial Vege-
tation. Cambridge University Press, Cambridge, U.K.
pp. 231-264.
and E H. Wagner. 1985. The Mojave, Sonoran and
Chihuahuan Deserts of North America. In: M. Evenari,
I. Noy-Meir, and D. W. Goodall (eds.), Hot Deserts and
Arid Shmblands, A. Elsevier Science Publishers, Am-
sterdam, pp. 105-202.
McClaran, M. P. and T. R. Van Devender (eds.). 1995. The
Desert Grassland. University of Arizona Press, Tucson,
AZ.
Merriam, C. H. 1890. Results of a biological survey of the
San Francisco Mountain region and desert of the Little
Colorado, Arizona. U.S.D.A., N. Am. Fauna 3: 1-136.
Miller, R. E and P. E. Wigand. 1994. Holocene changes in
semiarid pinyon-juniper woodlands. BioScience 44:
465-474.
Myers, R. L. and J. J. Ewel (eds.). 1990. Ecosystems of Florida.
University of Central Florida Press, Orlando, FL.
Odum, W. E. and C. C. Mclvor. 1990. Mangroves. In: R. L.
Myers and J. J. Ewel (eds.), Ecosystems of Florida. Uni-
versity of Central Florida Press, Orlando, FL. pp. 517-
548.
Osmond, C. B. and L. E Pitelka (eds.). 1990. Plant Biology
of the Basin and Range. Ecological Studies 80.
Springer-Verlag, Berlin.
Peet, R. K. 1988. Forests of the Rocky Mountains. In: M. G.
Barbour and W. D. Billings (eds.), North American Ter-
restrial Vegetation. Cambridge University Press, Cam-
bridge, U.K. pp. 63-101.
Samson, E and E Knopf. 1994. Prairie conservation in
North America. BioScience 44: 418-421.
Shreve, E and I. L. Wiggins. 1964. Vegetation and Flora of
the Sonoran Desert. Stanford University Press, Stan-
ford, CA.
Shugart, H. H., R. Leemans, and G. B. Bonan (eds.). 1992. A
Systems Analysis of the Global Boreal Forest. Cam-
bridge University Press, Cambridge, U.K.
Sims, P. L. 1988. Grasslands. In: M. G. Barbour and W. D.
Billings (eds.), North American Terrestrial Vegetation.
Cambridge University Press, Cambridge, U.K. pp. 265-
286.
Spicer, R. A. and J. L. Chapman. 1990. Climate change and
the evolution of high-latitude terrestrial vegetation
and floras. Trends Ecol. Evol. 5: 279-284.
Stuckey, R. L. and K. J. Reese (eds.). 1981. The Prairie
Peninsula—in the "shadow" of Transeau. Proc. of the
6th North American Prairie Conference, Ohio State
University, Columbus, OH, 1978. Ohio Biological Sur-
vey, Columbus, OH.
Suffness, M. (ed.). 1995. Taxol, Science and Applications.
CRC Press, Boca Raton, FL.
Transeau, E. N. 1935. The prairie peninsula. Ecology 16:
423-437.
Tucker, J. M. 1961. Studies in the Quercus undulata com-
plex. I. A preliminary statement. Am. J. Bot. 48: 202-
208.
Vankat, J. L. 1979. The Natural Vegetation of North Amer-
ica, An Introduction. Wiley, New York.
Setting the Goal: Modern Vegetation of North America 27
West, N. E. 1988. Intermountain deserts, shrub steppes, and
woodlands. In: M. G. Barbour and W. D. Billings (eds.),
North American Terrestrial Vegetation. Cambridge
University Press, Cambridge, U.K. pp. 209-230.
Wright, H. A. and A. W. Bailey. 1982. Fire Ecology: United
States and Southern Canada. Wiley, New York.
General Readings
Abrams, M. D. 1992. Fire and the development of oak
forests. BioScience 42: 346-353.
Chapin, E S. Ill and C. Korner (eds.). 1995. Arctic and
Alpine Biodiversity: Patterns, Causes, and Ecosystem
Consequences. Springer-Verlag, Berlin.
Cohn, J. P. 1996. The Sonoran Desert. BioScience 46:
84-87.
Gower, S. T. and J. H. Richards. 1990. Larches: deciduous
conifers in an evergreen world. BioScience 40:
818-826.
Gregory, S. V, E J. Swanson, W. A. McKee, and K. W. Cum-
mins. 1991. An ecosystem perspective of riparian
zones. BioScience 41: 540-551.
Joern, A. and K. Keeler. 1994. The Changing Prairie. Oxford
University Press, Oxford, U.K.
Kareiva, P. 1996. Diversity and sustainability on the prairie.
Nature 379: 673.
Mlot, C. 1990. Restoring the prairie. BioScience 40:
804-809.
Solbrig, O. T, E. Medina, and J. E Silva (eds.). 1996. Biodi-
versity and Savanna Ecosystem Processes: A Global
Perspective. Springer-Verlag, Berlin.
Van Cleve, K. et al. (+ 5 authors). 1983. Taiga ecosystems in
interior Alaska. BioScience 33: 39-44.
Whelan, R. J. 1995. The Ecology of Fire. Cambridge Univer-
sity Press, Cambridge, U.K.
Notes
1. For patterns of vegetation evident from the Advanced
Very High Resolution Radiometer, a sensor on U.S. Na-
tional Oceanographic and Atmospheric Administration
meteorological satellites, see Coward et al. (1985).
2. The most northerly forest in the world is a small
patch of conifers near 72° 30' N near Ary-Mas in northern
Russia (Spicer and Chapman, 1990).
3. Among the prominent organizations effective in the
preservation of these forests and other North American
biotas are the following:
The Nature Conservancy
1815 North Lynn Street
Arlington, VA 22209
Save-The-Redwoods League
114 Sansome Street
San Francisco, CA 94104
Sierra Club
730 Polk Street
San Francisco, CA 94109
 CHAPTER TWO
Cause and Effect
Factors Influencing the Composition and
Distribution of North American Plant
Formations through Late Cretaceous and Cenozoic Time
The arrangement of vegetation over the landscape is a
product of interactions between the environment, the eco-
logical characteristics of individual organisms, barriers,
dispersal potential, epidemic disease, anthropogenic influ-
ences, and the partially serendipitous factor of propagule
availability. Within the complex of environmental factors,
several are of special importance in tracing the history of
North American plant communities. They include climate;
plate tectonics as a mechanism for orogeny, volcanism,
land bridges, and terranes; and catastrophes. Each have nu-
merous interacting subcomponents, feedbacks, and ampli-
fiers, and although constraints of format make it necessary
to discuss these separately and sequentially, they are inter-
connected and pertubation of one affects the entire system.
Diagrams summarizing these factors are presented at the
end of the following sections. The diagrams are not in-
tended as models for, indeed, the single factor of climate
could be expanded into a component so vastly complex
that it would be counterproductive to a general summary.
Similarly, the hydrological cycle, which involves the
largest movement of any substance on Earth, cannot be
fully treated because a "systems" view of its role in influ-
encing climate is not available (Chahine, 1992) and the
roles of water vapor (a greenhouse gas) and cloud cover are
just now being quantified (Cess et al., 1995; Ramanathan et
al., 1995). Rather, the diagrams illustrate some of the fac-
tors and relationships discussed in the text and serve as a
reminder of the complex interactive nature of physical and
biotic events.
CLIMATE
Plants are limited in their ecological amplitude. Several
important corollaries follow from this observation; one of
the most fundamental is that changes in climate cause ex-
28
tinctions, promote evolution, and alter the range and habi-
tats of organisms. Because climate plays a central role in
the arrangement of modern communities (Gates, 1993;
Kareiva et al., 1993; Woodward, 1987) and in the develop-
ment and distribution of past assemblages (Brenchley,
1984; Crowley and North, 1991; Hecht, 1985a), reference to
some elements of general climatology is necessary for un-
derstanding the diversification, radiation, and reshuffling
of North American paleocommunities during the Late Cre-
taceous and Cenozoic.
Some General Aspects of Climate
Climate is the long-term weather of a region; it consists of
a plexus of dynamic, interrelated elements that are collec-
tively responsible for precipitation, temperature, and at-
mospheric and oceanic circulation1 (Barry and Chorley,
1992). Variations in these elements determine the different
regional climates, the seasonal changes in climate within a
region, and fluctuations in climate over time. The driving
force behind regional differences and temporal changes is
insolation, which is the amount of solar radiation or heat
reaching the Earth's surface.
The spatial distribution of heat varies because of the
curvature of the Earth: a given amount of heat is dispersed
over a greater surface area at the poles than at the equator
(Fig. 2.1). In addition, solar rays impinge on the Earth at a
more oblique angle near the poles, resulting in greater scat-
tering. At the equator the amount of solar energy received
at the top of the atmosphere is 311,520 cal/cm2/year, while
near the poles it is less than half this amount (130,300
cal/cm2/year; List, 1951, table 133). Consequently, any
change in insolation has a magnified effect on polar cli-
mates compared to equatorial climates, as evidenced by
the distribution and extent of glaciers.
Annual changes, or seasons, are a result of the tilt of the

Figure 2.1. Effect of the Earth's curvature on the distribu-
tion of equal amounts of heat from the sun at the equator
and at the poles, as represented by the widths of the two
sets of lines and the length of the dark bars, respectively.
Any change in incident radiation will have a magnified ef-
fect at the poles compared to the equatorial regions.
Earth during its orbit around the sun (Fig. 2.2). The tilt is
presently 23.5° from the orbital plane, so that 1 day of the
year, June 21 (the northern hemisphere summer), the sun's
rays are at the northernmost point where they strike the
Earth perpendicularly (the summer solstice, Fig. 2.2A).
This latitude is marked by the Tropic of Cancer at 23.5° N,
and it crosses North America at a point just south of Mate-
huala, Mexico, westward toward Mazatlan. On December
21 (the winter solstice), at a point in the orbit opposite the
Figure 2.2. Position and present orientation of the Earth relative to the sun at the summer and winter solstices. Presently,
the perihelion is January 4 and the aphelion is July 5. The solstices do not correspond to the perihelion and aphelion.
Cause and Effect 29
June 21 position (Fig. 2.2B), it is the southern pole that is
inclined toward the sun, and its rays reach the southern-
most point where they strike the Earth perpendicularly
(the southern hemisphere summer). This latitude is marked
by the Tropic of Capricorn at 23.5° S, and it crosses South
America between Rio de Janeiro and Sao Paulo westward
toward Antofagast in northern Chile. The term solstice
refers to the two times of the year when the perpendicular
rays of the sun reach these most distant latitudes and mo-
mentarily have no apparent northward or southward mo-
tion. The related event equinox (Latin for equal night)
refers to the two times of the year when the sun crosses the
equator (September 22 and March 21, Fig. 2.2C,D). There
are 12 h of day and night at the equator all year; at the
equinox in the midlatitudes there are about 15 h of day in
the summer and 9 h in the winter, and at the solstices in
the high latitudes about half of the days have nearly 24 h of
light and about half have nearly 24 h of darkness. Clearly,
any change in the tilt of the Earth's axis would affect not
only insolation and consequently the various elements of
climate, but also influence the fundamental physiological
process of photoperiod and induce changes in both the
composition and distribution of plant formations. As will
be seen later, the tilt of the Earth has changed over time,
and that has played an important role in the history of the
Earth's biota.
The annual progression of the direct rays of the sun over
the Earth's surface produces a differential warming of the
atmosphere that contributes to atmospheric circulation.
The equatorial region is the warmest latitudinal zone be-
cause it is at or near a right angle to the sun's rays through-
out the year. This warm equatorial air rises, cools, and
30 Late Cretaceous and Cenozoic History of North American Vegetation
loses its moisture; hence, the equatorial zone has high tem-
peratures, large amounts of annual rainfall, and tropical
vegetation. At about 30° N and 30° S latitudes, across north-
ern Mexico and southernmost Brazil, respectively, the
cooled dry air begins to descend, warming as it approaches
the Earth's surface, and creates the high-pressure systems
that characterize these latitudes. The result is low precipi-
tation, high evapotranspiration potential, and desert vege-
tation. This north-south circulation pattern of the equato-
rial and midlatitudes is known as the Hadley regime after
British lawyer George Hadley, who in 1735 wrote an early
work on the northeast trade winds. The Sonoran Desert ex-
ists, in part, because it is located beneath the descending
arm of the Hadley convection cell.
At about 60° N and 60° S latitudes, from the southern
coast of Alaska across to mid-Hudson Bay and through the
Drake Passage between South America and Antarctica, re-
spectively, the reheated air begins to rise again and at the
poles descends to contribute to the climatic environments
of the polar deserts. However, Hadley regimes are weaker
outside the equatorial zone and give way to a different pri-
mary form of atmospheric circulation. At the mid- to high
latitudes the circulation has less the form of a convection
cell, but rather it occurs as two to five superimposed east-
trending waves called the Rossby regime (after Swedish-
American meteorologist Carl Rossby). The separation be-
tween the Hadley and Rossby regimes is around 30° N and
30° S latitudes. The Rossby regime comprises waves of
long amplitude upon which are superimposed shorter
waves (Trewartha and Horn, 1980, fig. 5.3). The lows and
highs shown on daily weather maps in the high northern
and southern latitudes generally correspond to the troughs
and ridges of the Rossby regime. Within these westerly
winds (viz., from the west) there are zones of faster flowing
air. At an altitude of -12 km the average flow is 50-100
km/h (30-60 mi/h), but portions have air speeds of
150-300 km/h (90-180 mi/h). These cores of faster flow-
ing air are the polar and subtropical jet streams. Cyclones
(zones of low pressure with associated precipitation) and
fronts (zones of weather instability and steeply gradient
temperature at the interface between highs and lows) de-
velop beneath the jet streams and are carried eastward by
the westerly flow of air. The jet streams shift south in the
winter, bringing cold Arctic air into the midlatitudes, and
north in the summer. For example, a subtropical high near
40° N in July shifts to near 25° N in January. Consequently,
any factor that alters the nature and course of atmospheric
circulation and the seasonal meandering of the jet streams
will affect regional climates and vegetation. As will be seen
later, orogeny (mountain building) and ice sheets of sub-
stantial height are two such factors; they have also played
an important role in the history of the Earth's climate and
biota.
In addition to the vertical patterns of wind distribution,
there are horizontal patterns created by gravity, rotation of
the Earth, frictional drag between the air and the Earth's
surface, and atmospheric pressure, which is the weight of
air per unit area [14.7 lb/in.2, or 1013.3 mb (millibars) at
sea level]. Atmospheric pressure is the most fundamental
because it is the factor that initially sets air into motion.
Air flows from zones of high pressure at about 30° N and
30° S latitudes into zones of low pressure at the equator
and at about 60° N and 60° S latitudes (Fig. 2.3). Isobars
connect areas of similar atmospheric pressure, and the
present distribution of winter and summer highs and lows
as outlined by isobars is shown in Figs. 2.4 and 2.5.
The direction of the wind is affected at the surface by
the Coriolis Effect (after French physicist and mathemati-
cian Gaspard de Coriolis who described it in 1844), a de-
flecting force generated by the spinning of the Earth. This
force deflects winds circulating around highs and lows ei-
ther to the east or west to create the easterlies, westerlies,
and trade winds in the idealized pattern shown in Fig.
2.3B. Within regions of constant high pressure, air move-
ment is relatively static, forming a zone known as the
Horse Latitudes; these are two belts located over the oceans
(at about 30-35° N and S latitudes) that have high baro-
metric pressure, calms, and light variable winds (Fig.
2.3A).
In the Pacific, systems of high and low pressure (the
southeast and northeast trade winds) meet along a zone of
generally low pressure and cloudiness just north of the
equator called the intertropical convergence zone (ITCZ,
Fig. 2.3B). Its position is seasonally between 10° N (August
and September) and 3° N (February and March). The loca-
tion of the ITCZ is one factor reflecting global climatic
processes, as evidenced by events associated with a shift of
only a few degrees latitude.
El Nino-Southern Oscillation Events
Although the relationship is not well understood, a south-
ward displacement of the ITCZ to near or just below the
equator is an early signal for the development of an El
Nino-Southern Oscillation (ENSO) event (Diaz and Mark-
graf, 1992; Philander, 1983). The effects of the Southern
Oscillation component are reciprocal between the Pacific
and Indian Oceans (e.g., a change in wind velocity that
produces higher sea levels in one region corresponds to
lower sea levels in the other, hence, oscillation). An El
Nino is the eastern Pacific and coastal South American
component of an ENSO event that begins with an overall
warming and consequent atmospheric disturbance in the
equatorial Pacific. [For a color illustration based on satellite
thermography, see Holden (1992)]. Technically, it is defined
as the appearance of anomalously warm water along the
coast of Ecuador and Peru as far south as Lima (12° S), dur-
ing which a normalized sea surface temperature (SST)
anomaly exceeding 1 standard deviation occurs for at least
four consecutive months at three or more of the following
five coastal stations: Talara, Puerto Chicama, Chimbote,
Isla Don Martin, and Callao [Quinn et al., 1987; Scientific

Figure 2.3. Schematic distribution of idealized high- and low-pressure systems and corresponding
wind directions.
Committee on Oceanic Research (SCOR), 1983]. The name
was coined by Peruvian fishermen because the event gen-
erally occurs around the Christmas season. The occurrence
of El Nifios varies, but they commonly appear every 3-7
years. They begin in the northern hemisphere spring, last
from 1-2 years, and alternate with normal or cooler than
normal episodes called La Ninas. There is an initial warm-
ing of equatorial eastern Pacific waters from 1 to 2°C to 5°C
or more above average that displaces the track of the sub-
tropical jet stream and weakens the tropical easterly trade
winds off the western coast of South America. This allows
warm ocean currents to flow more strongly eastward, dis-
rupting the cold Humboldt Current, increasing atmospheric
moisture, and producing climates not only atypical for
western South America but also for distant parts of the
Figure 2.4. Distribution of January high- and low-pressure systems. Adapted from Trewartha and Horn (1980) and
based on Mintz and Dean (1952).
Cause and Effect 31
world, including North America. Among the climatic ef-
fects of a typical ENSO event are droughts in Australia, In-
donesia, and southern Africa, and the bleaching of coral
reefs due to a westerly shift in wind direction that lowers
sea level in the west and raises it in the east (Glynn and de
Weerdt, 1991). There are excessive rains along the west
coast of South America and in the southwestern United
States (viz., the Texas to southern California floods of early
1992, flooding in the midwest in 1993), unusually cold
winters in northeastern North America, and unusually dry
warm winters in the midcontinent region and the west. For
example, during non-El Nino years the snowfall in Cleve-
land, Ohio, is -100 in./yr, while during an El Nino, it is
half that amount. The 1982-1983 event (Glynn, 1990;
Table 2.1), one of the most intense oceanographic and cli-
32 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 2.5. Distribution of July high- and low-pressure systems. Adapted from Trewartha and Horn (1980) and based on
Mintz and Dean (1952).
matic disturbances in the Pacific in the past 100 years, was
anomalous because the effects were evident in May rather
than in the fall and occurred in the central and western Pa-
cific rather than along the west coast of South America.
Rainfall increased sixfold to 2540 mm (100 in.) over an 8-
month period, converting the coastal desert of Ecuador and
northern Peru into a lake-dotted grassland. The El Nino of
1997-1998 is predicted to be comparable in intensity to
the 1982-1983 event. A strong El Nino produces a global
warming of-0.1-0.2°C. The cause(s) of ENSO events are
not known, but there is evidence to suggest a relationship
with variations in the rate of the Earth's rotation that may
produce cyclic changes in the atmospheric and oceanic
distribution of heat and moisture (Hide and Dickey, 1991).
The economic impact of an El Nino is evident by the ex-
tensive media coverage, such as CNN's Financial Market
Reports on the Web during September-October, 1997:
"Markets weather El Nino—onset of phenomenon blows
more volatility into commodities trading; El Nino hits
nickle producer; El Nino hits coffee crop; Making El Nino
Table 2.1. Chronology of strong (S) to very strong (VS)
ENSO events in nineteenth and twentieth centuries.
1803-1804 (S+) 1877-1878 (VS) 1925-1926 (VS)
1814 (S) 1884 (S+) 1932 (S)
1828 (VS) 1891 (VS) 1940-1941 (S)
1844-1845 (S+) 1899-1900 (S) 1957-1958 (S)
1864 (S) 1911-1912 (S) 1972-1973 (S)
1871 (S+) 1917 (S) 1982-1983 (VS)
Adapted from Quinn et al. (1978, 1987); see also Mass and Portman
(1989).
into dinero—Small firms still profit from storm pattern."
The importance of El Ninos to short-term weather is fur-
ther witnessed by the current 40-nation effort to establish
an international research center to forecast and predict the
impact of El Ninos on climate (Macilwain, 1995).
Eventually it may be possible to detect El Ninos in times
preceding historical records. Long-chain organic molecules
such as alkenones enter the sedimentary record as bio-
chemical fossils. In coccolithophorid algae such as Emilia-
nia huxleyi, the degree of alkenone unsaturation has been
shown to be temperature dependent (increasing unsatura-
tion as the water cools). If under favorable conditions
alkenone patterns are preserved in sediments for long
periods of time, ocean temperature changes could be re-
solved on a near-annual basis (Kennedy and Brassell,
1992). The irregularity (chaos) in the appearance and dur-
ation of El Ninos is also being investigated (Jin et al.,
1994; Tziperman et al., 1994), but presently it cannot be
explained.
THE IDEALIZED PATTERNS of temperature, rainfall, and wind di-
rection are modified locally by three geographic features
of the Earth's surface: topography, large bodies of water,
and large masses of land. Large-scale topographic effects
are induced by the presence and orientation of mountain
systems and plateaus. The windward (west) side of the
western mountains of North America receives moisture
from the westerlies off the Pacific Ocean (orographic pre-
cipitation), while a rainshadow is created to the east (lee-
ward) side. Rainshadows occur to the east of the Sierra
Nevada, contributing to the development of the arid Basin

and Range Province vegetation, and to the east of the Rocky
Mountains, contributing to the development of the High
Plains and Great Plains woodland and grassland vegeta-
tion. In addition to these regional effects, topography on a
smaller scale is important to local climates. The eastern
suburbs of Cleveland, Ohio (e.g., Lorain), receive ~1000
mm (40 in.) of snow each year. Fifty miles to the east at
Chardon, the annual snowfall is 2690 mm (106 in.) and
the seasonal maximum is 4100 mm (161 in.; Schmidlin,
1989). Winds circulating clockwise around Canadian high-
pressure systems bring cold Arctic air across Lake Erie,
which is shallow and normally does not freeze until Febru-
ary. The lake provides a source of moisture and a warmer
surface than the frozen land. Snow is generated when the
warmed, moisture-laden winds off the lake cross back onto
colder land on the southern shore. This system combines
with an elevational difference of 230 m (690 ft) to produce
up to 4 times more snow in Chardon than in Lorain, which
are at the same latitude and only 50 mi apart.
Another local effect of topography, in combination with
other factors, is the generation of the Santa Ana winds of
southern California. The winds develop around a high-
pressure anticyclone system, flow over the Sierra Nevada,
and gain heat on the western side as they subside. Addi-
tional heat is gained from crossing the warm lower eleva-
tions. The velocity of the winds increases during the fall
and winter months when clockwise circulating winds to
the north converge either with counterclockwise circulat-
ing winds to the south or with less intense highs. These
pulses of strong winds gain additional speed as they are
funneled through the canyons of the Sierra Nevada. The re-
sult is a strong, hot, dry, dusty winter wind that contributes
significantly to the extent and intensity of forest fires in
southern California, to high evapotranspiration, and to the
maintenance of arid shrubland/chaparral-woodland vege-
tation. Prior to significant elevations in the Sierra Nevada
(-10 Ma), these seasonal winds did not develop, which
was one of the altered boundary conditions that sustained
more mesic vegetation in the region.
Large bodies of water buffer surrounding terrestrial
habitats against extreme annual temperature fluctuations.
Oceans receive about 80% of the total solar radiation
reaching the Earth's surface. Marine waters have a specific
heat of 0.93 cal/g/°C, compared to that of land with 0.19-
0.60 cal/g/°C, and yield heat more slowly than land. Thus,
oceans serve as vast reservoirs of heat that maintain cli-
matic stability (Berggren 1982). The mean annual temper-
ature (MAT) of marine waters ranges from ~0°C at the
ocean bottom to ~28°C at the equatorial surface, and it usu-
ally varies annually by only a few degrees. In contrast,
MAT on land ranges from approximately -40 to 30°C and
can vary annually by 50°C. The extensive epicontinental
sea on the North American continent during the Late Cre-
taceous was a significant factor in producing the equable
maritime climates and low thermal gradients characteristic
of that period.
Cause and Effect 33
In addition, oceans and large lakes are responsible for
sea and land breezes that can modify local climate over
short distances. At sunrise land surfaces warm faster than
adjacent bodies of water and a rising flow of air is gener-
ated. Cool air from over the water is drawn inland and may
decrease average daytime temperatures by several degrees.
For example, the daily mean maximum July temperature at
Milwaukee, Wisconsin, on the south shore of Lake Michi-
gan is 25°C (77°F) while at Madison, which is 125 km (75
mi) inland, it is 28°C (82°F).
Large masses of land alter the idealized global zonation
of climate by setting the broad pattern of ocean circulation
and by producing extremes of temperature and precipita-
tion characteristic of interior continental climates. The
withdrawal of the epicontinental sea during the Cenozoic
promoted greater seasonal temperature extremes (conti-
nentality) in the interior of North America.
Another consequence of wind patterns induced by inso-
lation regimes are ocean surface currents (Fig. 2.6). High-
and low-pressure systems, prevailing winds, the position
and configuration of the continents, and the spinning mo-
tion of the Earth produce gyres—great circulating systems
of surface ocean water (Pedolosky, 1990). A factor further
contributing to the motion of marine water is the difference
in salinity in different parts of the ocean. There is less
oceanic rainfall under the descending branch of the Hadley
cell, with consequent higher salinity and greater evapora-
tion from marginal seas (Price et al., 1993). The center of
the gyres marks the position of semipermanent, seasonally
variable, high-pressure systems; the spinning of the Earth
produces the familiar clockwise circulation in the northern
hemisphere and the counterclockwise motion in the south-
ern hemisphere about these columns of air, and these mo-
tions are transferred to the surface of ocean waters. The
southern limits of the northern hemisphere gyres are de-
fined by the seasonally migrating ITCZ, the northern limits
by the high latitude easterlies, and the east-west limits by
the location and configuration of the continents (Berggren,
1982). The circulating motion of the gyres pulls warm
water from the equatorial region and cold water from the
poles, contributing to the global distribution of heat, and is
responsible for many features of climate, and especially
noticeable along continental coastlines. Where cold cur-
rents flow along the coast, such as the Humboldt or Peru
Current along the western coast of South America, the cool
dry winds that blow onto the land extract moisture from
the atmosphere and contribute to the formation of the dry
climate and vegetation of the Chilean Atacama Desert.
Studies of diatomites from western South America show
that upwellings developed at -39 Ma, indicating that a
proto-Humboldt Current was in existence by that time. Up-
wellings are an upward movement of cold, nutrient-rich
bottom waters, and they appear during periods of in-
creased circulation of surface waters. These increases
occur with the opening and closing of land bridges, such as
the Drake Passage between South America and Antarctica,
34 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 2.6. Principal ocean currents affecting North American climates: (1) East Gulf loop, (2)
Florida current, (3) North Atlantic drift, (4) east Greenland current.

and across the Isthmus of Panama. Upwellings produce
blooms of marine organisms in response to increased nu-
trient levels, and these blooms are preserved in ocean sed-
iments. The South American diatomites correspond to a
drop in sea level at -40 Ma, which probably represents
early glaciation on Antarctica and the initiation of cold
South Atlantic Deep Water (SADW) production.
Where the current is warm, such as the Gulf-Agulhas
current flowing into the western North Atlantic Ocean, the
effect on local climates is quite different. About 30 million
m3 of seawater are transported through the Straits of
Florida into the North Atlantic each second (annual mean).
The Agulhas Current of the Indian Ocean flows westward
around the tip of Africa and feeds considerable amounts of
warm water into the Gulf Stream (Fig. 2.6; Bird, 1991). The
average temperature of this water is 17.4°C. These Western
Boundary Currents (WBCs) displace climatic zones pole-
ward by 5-10°. The presence of tropical zooxanthellate
corals (those with dinoflagellate symbionts) in Bermuda
(latitude 33° N) is one biogeographic consequence of the
current. Farther north, westerly winds blowing across
these waters are warmed; by the time they reach the coast
of England [average cold month temperature of 5.5°C
(41°F)], this sea to air temperature exchange produces a
maritime climate very different from the continental cli-
mate in Labrador [average cold month temperature of
-25°C (-13°F)], which is at the same latitude (51.5° N). The
North Atlantic current also receives an influx of highly
saline water from the Mediterranean which increases the
density of the water. The cooled, saline-dense current be-
gins to sink in the Labrador and Nordic Seas to form North
Atlantic Deep Water (NADW) that flows southward at
depths greater than 1000 m, eventually reaching the east
coast of South America. Compared to the 17.4°C tempera-
ture of the north-flowing surface water, the returning
south-flowing water has an estimated average temperature
of 5.4°C (Covey and Barren, 1988). At the tip of South
America it trends east along the Antarctic coast, then north
to join the warm equatorial Agulhas Current that feeds
back into the Gulf Stream. This conveyor belt system has a

period of -1600 years. Intermittent NADW production
began in the early Oligocene, suggesting cooler climates
and further glaciation on Antarctica, and reached a pattern
typical of modern times between 10 and 7 Ma. The Isthmus
of Panama began closing about this time with the forma-
tion of a submerged sill that deflected and increased the
flow of the Gulf Stream. The Isthmus was essentially
closed by -3.5 Ma, at which time upwellings increased
along the western Atlantic coasts. In the present configura-
tion of the continents, a part of the Agulhas Current is de-
flected northward by the projection of Brazil (Fig. 2.6) and
it augments heat transport by the warm Gulf Stream. A
slightly eastward position of South America, as during the
Tertiary, would cause a greater southern deflection, which
would alter ocean circulation and the global distribution of
heat. Seaways and land bridges across the North Atlantic
Ocean and North Pacific Ocean, through Panama, and be-
tween South America and Antarctica were altered during
the Cenozoic. These changes had a significant effect on
ocean circulation, climate, and the history of the Earth's
biota.
In the North Pacific less heat transport occurs, and the ef-
fect on regional temperatures is quite different from that in
the North Atlantic. Bodo, Norway, has a January average tem-
perature of-2°C and a July average of 14°C. In contrast, the
January and July values at Nome, Alaska, at a similar latitude
but off the Pacific Ocean are -15 and 10°C, respectively.
The temperature, precipitation, wind, and ocean cur-
rents generated by insolation and modified by the position
and configuration of continents and their topography pro-
duce the climates typical of different regions of North
America. Within each component of climate the parame-
ters most important to the composition and distribution of
plant formations are the extremes, as in temperature and
precipitation, because it is the extremes that interact with
the ecological tolerances of individual organisms to deter-
mine potential distributions. The MATs of Des Moines,
Iowa, and Oakland, California, are -11 and ~13°C, respec-
tively; but the annual range at Oakland is 9.5°C, and at Des
Moines it is 30.4°C, and the two regions support com-
pletely different potential natural vegetation. Average sea-
level temperatures for January and July in North America
are shown in Figs. 2.7 and 2.8, respectively, and precipita-
tion for the same months are shown in Figs. 2.9 and 2.10,
respectively. The values reflect the extremes in yearly vari-
ations with which all organisms in a given region must
contend and be physiologically capable of surviving. It is
interesting to note that present mean seasonal extremes are
greater than the MAT difference between the glacial and
interglacial phases of the Pleistocene (Crowley and North,
1991). The interaction of climatic extremes and ecological
tolerances, modified by other more local factors of the en-
vironment and depending on propagule availability, pro-
duce patterns in the distribution of plant species. The par-
ticular patterns determined by present conditions are
shown on various vegetation maps (Barbour and Billings,
Cause and Effect 35
1988; Barbour and Christensen, 1993; Kiichler, 1964) and
on the summary diagram in Fig. 1.1.
It is evident from the preceding discussion that these
patterns are tenuous and that a change in any one of the
numerous factors that determine climate will, in turn, in-
fluence the composition and arrangement of plant forma-
tions and associations. It may seem that the changes evi-
dent over geologic time could be attributed primarily to
known fluctuations in the amount of radiation emanating
from the sun because insolation is the source of energy de-
termining the Earth's climate. At a 150 million km (93 mil-
lion mi) mean distance from the sun, a surface oriented
perpendicular to the sun receives -2.0 cal/cm2/min. This
value is known as the solar "constant." In fact, there have
been fluctuations in the amount of heat coming from the
sun, and this may have been a factor in determining pat-
terns of temperature during the Cenozoic (Haigh, 1996;
Hoyt and Schatten, 1997; Kerr, 1996a, General Readings).
However, changes in solar heat alone do not sufficiently ex-
plain the Earth's climatic history; there is presently no evi-
dence to document that it has fluctuated in a fashion corre-
sponding to the patterns of extinctions, evolutionary
peaks, or the migrational history of the Earth's biota. Peri-
odically, solar flares of ultraviolet radiation are trapped in
the Earth's magnetic field and are concentrated above the
north and south magnetic poles. At an elevation of about
100 km (60 mi), these charged particles collide with atmo-
spheric gases to produce the aurora borealis and aurora
australis. Although such variations in solar radiation are
visually spectacular, they do not appear to cause any sig-
nificant changes in global climate. [For a possible correla-
tion between sun spot activity and short-term climate fluc-
tuations, see Flam (1993), Friis-Christensen and Lassen
(1991), Kerr (1995, 1996b)]. As noted by Covey (1996), "It
would be foolish to assume that historical solar luminosity
changes are insignificant. We should not, however, make
the opposite error and jump to the conclusion that they ex-
plain all—or even most—of the climate variations of the
past few centuries." If fluctuations in solar radiation alone
do not correspond in detail to the history of the Earth's cli-
mate and biota, then other factors affecting insolation must
be involved, such as variations in the position and orienta-
tion of the Earth relative to the sun. These variations have
occurred over time, and they are a driving force behind
global climatic change (Berger et al., 1984).
Milankovitch Variations
The Serbian astronomer and mathematician Milutin Mi-
lankovitch knew from previous studies (e.g., Croll, 1875)
that cyclic changes in insolation occurred as a result of
variations in the distance and orientation of the Earth rela-
tive to the sun. The variations derive from fluctuations in
the gravitational pull on the Earth's equatorial bulge by the
moon, sun, and planets in the solar system, notably Jupiter
and Saturn, depending on their positions and distances
Figure 2.7. Mean January temperatures for North Amer-
ica. Based on the Climatic Atlas of North and Central
America—Maps of Mean Temperature and Precipitation
(World Meteorological Organization, 1979). Scale
1:10,000,000.
10 50 100 200
Figure 2.9. Mean January precipitation for North Amer-
ica. Based on the Climatic Atlas of North and Central
America—Maps of Mean Temperature and Precipita-
tion (World Meteorological Organization, 1979). Scale
1:10,000,000.
Figure 2.8. Mean July temperatures for North America.
Based on the Climatic Atlas of North and Central Amer-
ica—Maps of Mean Temperature and Precipitation
(World Meteorological Organization, 1979). Scale
1:10,000,000.
10 50 100 200
Figure 2.10. Mean July precipitation for North America.
Based on the Climatic Atlas of North and Central Amer-
ica—Maps of Mean Temperature and Precipitation
(World Meteorological Organization, 1979). Scale
1:10,000,000.

relative to the Earth. The spatial relationship between the
Earth and sun varies in three ways. One is in the tilt of the
Earth's axis or obliquity. The tilt is presently at 23.5° (Fig.
2.2), but it varies between 22.1° and 24.5°. For example, at
9 Kya the tilt was 24.2°. Obliquity affects the distribution of
heat over the Earth's surface, particularly the insolation dif-
ferential between the high and low latitudes (large effects
at the poles, less toward the equator), and the intensity of
the seasons. An increase in the tilt results in greater sea-
sonal contrast (warmer summers and colder winters). For
the northern hemisphere the maximum difference can be
~14°C (+7°C in July and -7°C in January). The period of the
tilt variation is -41,000 years.
The second variation is in the eccentricity of the ellipti-
cal orbit (Clemens and Tiedemann, 1997). If the Earth's
orbit around the sun were a perfect circle, the eccentricity
would be zero; if one dimension were twice as long as the
other, the eccentricity would be 0.866. At present it is
0.017, and it varies between 0.005 and 0.06. The perihelion
is the point where the Earth is nearest the sun, which is
now reached on January 3 (Fig. 2.2); the aphelion is the
point where it is furthest away, which is now reached on
July 4. About 9 Kya the perihelion was July 30. Presently
the Earth receives -6% more heat at perihelion than at
aphelion, while at maximum eccentricity the difference is
27%. Changes in eccentricity alternately intensify the sea-
sons in one hemisphere and moderate them in the other.
The period of the eccentricity variation is -100 ka, and
there is a secondary period at -400 ka.
The third variation is in the circular movement or pre-
cession of the polar axes. As the Earth spins there is a slight
movement in the direction toward which the polar axes
point, so that over time they would inscribe a circle if pro-
jected onto a distant background. The torque exerted on the
Earth by the moon and planets accounts for about two-
thirds of the precession rate and that exerted by the sun ac-
counts for the remaining one-third. At present the north
pole points toward a star that is called the pole star or Po-
laris (the end star in the handle of the Little Dipper). In
about 12 Kya it will point toward Vega. This precession is
a superimposed, second-order factor that reinforces the in-
tensity of the seasons, lessening it in one hemisphere
(presently the northern hemisphere, with comparatively
mild winters) and increasing it in the other (presently the
southern hemisphere, with comparatively cold winters). In
contrast to the tilt variable, the effect of changes in preces-
sion is small at the poles (-0.5%) and larger toward the
equator. It takes -23 Ky for the Earth's axis to inscribe a full
imaginary circle as it processes, and a secondary cycle
takes -19 Ky.
The three kinds of variations just described are eu-
phemistically referred to as tilt, stretch, and wobble. Mi-
lankovitch (1920, 1930, 1941) calculated their individual
and combined effects on insolation for each latitude over
time. He concluded that when periods of low solar radia-
tion coincide with minimal seasonal ranges in temperature
Cause and Effect 37
(viz., greater summer accumulation of winter snow), gla-
cial conditions develop. The combined effects of the three
variations can vary summer insolation in the northern
hemisphere by a maximum of -20%. His calculations pre-
dicted glaciations every 100 Ky. When high insolation is
matched with maximum seasonal ranges in temperature
(viz., greater summer melting of winter snow), the glacial
periods should end by these calculations for 10-12 Ky.
Both cycles have been confirmed by a variety of marine
and ice-core evidence (Imbrie et al., 1992, 1993; Schwar-
zacher, 1993; see following section). Milankovitch's origi-
nal calculations have been reviewed by Berger (1980).
At different times in Earth history and at different lati-
tudes each of the variations have dominated solar-
controlled climatic patterns. Before 2.4 Ma the dominant
cycles were 23 Ky and 19 Ky precessions, while after 2.4
Ma the 41 Ky obliquity cycle dominated. During the inter-
val between 800 and 620 Kya, climatic systems in the high
latitudes underwent a change from primarily 41 Ky obliq-
uity forcing to the primarily 100 Ky eccentricity forcing of
the present (Fischer, 1982; Shackleton et al., 1990).
In the years following Milankovitch's work there was
limited acceptance of the theory for two principal reasons.
One was that the scale of the theory, both temporal and
spatial, made it impossible to test directly or to establish a
cause and effect relationship with climatic changes and the
Pleistocene glaciations. With the technology of the day it
was also impossible to determine whether other cyclic
events evident in the geologic record, such as repeating cy-
cles of sedimentation, corresponded to periods predicted
by the Milankovitch calculations. Supporting evidence
was accumulating, however (Broecker and van Donk, 1970),
and it is somewhat ironic that strongest support for the the-
ory, which was originally proposed to explain continental
glaciations, should ultimately come from the sea.
18
O/16O Ratios
In 1947 Harold Urey suggested that the relative amounts of
oxygen isotopes in the mineral walls of marine organisms
were a function of the ocean water temperature and more
18
O was taken up as the water cooled (see also Epstein et
al., 1951; Urey et al., 1951). Various marine organisms such
as molluscs, Coccolithophorae, and foraminifera synthe-
size their calcium carbonate walls from molecules avail-
able in the ocean water; Emiliani (1955) demonstrated that
the shell chemistry of modern foraminifera does indeed re-
flect the oxygen isotope ratio of the ambient water. (One
molecule of seawater per 500 has the 18O molecule.) The
shells become incorporated into bottom sediments accu-
mulating at depths less than ~4 km, the calcite compensa-
tion depth, and they preserve a record of fluctuating
18Q/16Q ratios. At greater depths the shells are dissolved in
the unsaturated water. In 1966 the research vessel Chal-
lenger began to retrieve cores from all the ocean basins,
and in 1968 the cores formed the basis for the DSDR An ac-
38 Late Cretaceous and Cenozoic History of North American Vegetation

count of these historic and scientifically important voyages

is provided by Hsu (1992), and the project continued as the
ODE
The cores were analyzed for a variety of climatic indica-
tors, including 18O and 16O. The waters of the world's
oceans mix on a time scale of ~1600 years (ventilation
rate), so their isotopic composition is relatively uniform
and results can be correlated among the different ocean
basins (Hecht, 1985b). A 30-year trend in the 18O composi-
tion of modern precipitation and surface air temperature
confirmed the relationship (Rozanski et al., 1992), and the
isotope ratios were calibrated to real temperature values by
comparison with modern analogs and through aquaria ex-
periments. A factor that complicates the derivation of tem-
perature estimates from the 18O/16O data is that the ratio
also varies with salinity and ice volume (Chapter 3). Nev-
ertheless, sequences of cool-warm intervals are clearly ev-
ident in the ocean cores, and when the timing of magnetic
pole reversals and extrapolations were used to date the cy-
cles, they revealed periodicities of 23, 41, and 105 Ky
(Hays et al., 1976; see also Shackleton et al., 1990); the 100
Ky cycle was the primary one (Fig. 2.11).
In addition to DSDP and ODP core segments with pat-
terns of high versus low 18O concentrations, support for the
Milankovitch theory has come from several other sources.
An early source was from patterns of sea-level change pre-
served as erosion terraces on emergent Acropora palmata.
coral reefs in Barbados. The reefs grow at the water surface
so that a rise in sea level, as during an interglacial, results
in new growth to keep pace with rising ocean waters. The
top of an ancient reef thus marks a high stand in sea level
and, by inference, a period of warm climate. When the sea
level is lowered, as during glacial periods and cold cli-
mates, a terrace is eroded into the coral mass. Eighteen
major terraces are preserved on the island. By using ura-
nium series dating techniques, Broecker et al. (1968) and
Mesolella et al. (1969) reported high stands of the sea at 82,
103, and 122 Kya that were roughly in agreement with the
tilt and precessional orbital variations calculated by Mi-
lankovitch for the lower latitudes (45° N at 88, 110, and
132 Kya; the "Barbados sea-level model"; Ku et al., 1990).
Changes in North Atlantic Ocean currents also provided
a clue to the timing of glacial events. Ruddiman and Mcln-
tyre (1976, 1984) traced the course of the Gulf Stream by
analyzing cores for shifts in the distribution of tempera-
ture-sensitive marine species. During the interglacials the
Gulf Stream flowed northeast toward Great Britain, while
during glacial times it flowed more easterly toward Spain.
The fixed point of the gatelike swings was near Cape Hat-
teras, North Carolina. Comparison of the swings with the
magnetic time scale revealed a cycle of 100 Kya.
There are also indications of similar cyclic events in
much older deposits (Berger et al., 1984, 1992). The New-
ark Basin in northern New Jersey is a remnant of a series of
filled rift valleys about 200 Ma old that developed in the
Late Triassic as North America separated from Africa. Indi-
Figure 2.11. The marine oxygen
isotope record through the past
800,000 years. Stippling indicates
periods of cooling temperatures.
Matuyama-Brunhes marks a re-
versal in the Earth's magnetic pole.
Adapted from Williams et al.
(1993) and based on Imbrie et al.
(1984). Reprinted with the permis-
sion of M. A. J. Williams and
Kluwer Academic Publishers.
vidual units about 6 m thick are recognizable on the basis
of sediment type, texture, fossils, and other organic matter
content. Van Houten (1964) suggested these cycles might
correspond to periods of climatic change induced by or-
bital variations, but in the absence of accurate age determi-
nations it was not possible to establish a close association.
Recently, Olsen (1986) and Olsen and Kent (1996) applied
radiometric techniques to the sediments and found ages of
25, 44, 100, and 400 Ky, which compare to orbital periods
of 23, 41,100, and 400 Ky. Fourier analysis of Mesozoic cli-
mate proxy data and DSDP cores show similarities be-
tween rates of sedimentation, accumulation of calcareous
marine microfossils, and orbital cycles in Middle Creta-
ceous deposits in Italy (100 Ma; 100 and 400 Ky cycle),
Late Triassic deposits near Lyme Regis in southern coastal
England (200 Ma; 40 Ky cycle), and the Late Cretaceous of
the South Atlantic (DSDP site 516F, 30° S; precessional
cycle of -23 Ky; Park et al., 1993). Other examples are
given by Arthur and Garrison (1986) and Berger et al.
(1984).
Although the Milankovitch theory is now generally ac-
cepted, the correspondence between cyclic climatic events
and periods of the orbital cycles has been challenged.

Winograd et al. (1988) studied 18O variations in calcite de-
posited from ground water on the walls of a submerged
open fault called Devils Hole near Ash Meadows, Nevada,
which is 120 km northwest of Las Vegas. The deposits pre-
serve a continuous 250 Ky record of climatic change in the
Great Basin between the Middle and Late Pleistocene
(-310-50 Kya). The general sequence of events is similar
to that in ocean sediments and ice cores, but the timing is
different. The dates from Devils Hole suggest that the pre-
vious interglacial [Termination II in the terminology of
Broecker and van Donk (1970)] began about 147 Kya or ear-
lier; other records, in agreement with the timing of the or-
bital cycles, place the date at about 127 Kya. An earlier in-
terglacial was dated at 272 Kya, while the marine and
ice-core date is 244 Kya.
In the midst of these discussions, additional support for
the Milankovitch theory came from new dates obtained
from the coral reefs that corrected for inaccuracies due to
diagenesis (Gallup et al., 1994). Dating of Termination II
from Barbados coral by a refined alpha particle counting
technique (Rea et al., 1985) yielded an approximate age of
126 Ky; a new method based on thermal ionization mass
spectrometry (TIMS) yielded an age between 122 and 130
Ky. Both dates are in agreement with those from ice cores
and marine sediments for Termination II and are consistent
with the timing of the Milankovitch cycles. Spectral analy-
sis of layers of eolian (wind blown) dust particles in a
DSDP core from the central North Pacific revealed fluctua-
tions associated with the comparatively moist interglacial
and drier glacial climates, and these are also consistent
with dates from the marine realm for Termination II. Sub-
sequent peaks in the Barbados coral curve were dated at
23, 41, and 104 Kya. Supporters of the Milankovitch varia-
tions have suggested that dates from Devils Hole are prob-
ably inaccurate because of contamination by 230Th (Ed-
wards and Gallup, 1993).
Longer and better dated cores from Devils Hole later
provided a 500 Ky record (between 560 and 60 Kya; Wino-
grad et al., 1992), and uranium-thorium dates of coral reefs
from New Guinea (135 Ky) and in the Bahamas (132 Ky)
suggest an end to glacial cycles earlier than predicted by
the Milankovitch model. Imbrie et al. (1993) believed there
were flaws in the dates and interpretation of the Devils
Hole material because calculations show that direct appli-
cation of the Devils Hole chronology to ocean cores would
result in implausible rates of marine sedimentation (see re-
sponse, Ludwig et al., 1993). U-Th dates of marine sedi-
ments for the last two interglacial periods (120-127 and
189-190 Ky) are in accord with variations in the Earth's
orbit as a fundamental cause of Quarternary climatic
change (Slowey et al., 1996).
The latest contribution to these discussions has been
application of U235 /Protactinium231 dating to the Devils
Hole carbonates that confirms the original date for warm-
ing at ~140 Kya (Edwards et al., 1997). The view is now
emerging that because the Barbados coral data and sea-
Cause and Effect 39
level changes are in agreement with the Milankovitch cy-
cles, they are the principal cause for global-scale climatic
changes during the Quaternary. Confirmation of the Devil
Hole dates are now suggested to reflect the local climatic
cycles of the southwestern United States.
In addition to these problems, some other challenging
questions have not been fully answered. In particular, it is
unclear why orbital variations should produce extensive
glacial-interglacial cycles during the last -2.4-1.6 m.y.
but not earlier. It is possible that overriding high CO2 con-
centrations in the Mesozoic and Paleogene had waned by
the Neogene to the point that the Milankovitch variations
could be expressed as a primary forcing mechanism of cli-
mate. Another problem is that a change in the shape of the
orbit (100 Ky cycle) alters insolation by about 0.5%, while
tilt and precession can change it by as much as 14-27%.
This raises the question of why major climatic cycles
should follow the pattern of least heat fluctuation. Clearly
some amplification of the 100 Ky cycle is needed, and ex-
planations are now being formulated. They include pat-
terned overlap between equinox and perihelion (Crowley
et al., 1992), CO2 concentration (see later section), varia-
tions in the frequency of the obliquity cycle (Liu, 1992),
and possibly variations in the accretation rate of interplan-
etary dust (-10 K tons of extraterrestrial material fall on the
Earth each year). Variation in this amount may result from
changes in the inclination of the orbital plane that follow a
100 Ky cycle (see papers mentioned in Brownlee, 1995).
With regard to the change from one dominant forcing
mechanism to another, Imbrie et al. (1993) suggest that ice
sheets, resulting from precession and obliquity forcing,
upon exceeding a critical size, act as negative feedback that
imposes the slower eccentricity cycle of change.
Another problem is that northern and southern hemi-
sphere glaciations are nearly synchronous (Schneider and
Little, 1997), but because of the Earth's tilt they should al-
ternate. This may be due partly to the facts that Antarctica
is surrounded by a circumpolar ocean current and it is a
smaller continent unjoined to other land masses (less con-
tinentality). Thus, it may have a higher threshold for the
formation of glaciers. Models are being developed that ex-
plore the role of cooler equatorial oceans in both amplify-
ing orbitally-induced changes in seasonality and in the in-
terhemispheric correlation of glacial events (Bard et al.,
1997; Beck et al., 1997; Curry and Oppo, 1997; Webb et al.,
1997). Finally, climates can change abruptly, within a few
to several hundred years, as shown by analyses of ice cores
from Greenland; abrupt changes are not explained by the
gradual and steady pace of the orbital cycles. It is clear that
other processes that produce more modest and rapid shifts
in climate must be superimposed on the Milankovitch
variations. Witnessing the debates and the fine-tuning of
significant unifying concepts imparts to each period of sci-
ence its own uniqueness and special importance. Emerging
data from the ice cores and refinements in the orbital the-
ory of glaciations are among those that mark this period. It
40 Late Cretaceous and Cenozoic History of North American Vegetation
is a challenging and fascinating time to be contemplating
Earth history and biotic response because existing tech-
nologies are being pushed to their limits, new methodolo-
gies are being developed, and broad answers to immensely
complex questions are just being formulated.
Greenland Ice-Core Record
During the last deglaciation there was an abrupt cooling
event in Europe between -11 and 10 Kya called the
Younger Dryas (Broecker et al., 1988; Severinghaus et al.,
1998). Pollen and spore diagrams revealed that forests that
had begun to recolonize western Europe were replaced by
tundra vegetation, while in eastern North America the veg-
etation mostly continued a trend toward deciduous forest.
Such rapid shifts in climate are now being confirmed
through the European GRIP (Dansgaard et al., 1984, 1993;
GRIP Members, 1993; Grootes et al., 1993; Johnsen et al.,
1992; Paterson and Hammer, 1987; Taylor et al., 1993a,b),
and the U.S. GISP2. Annual layers are present in the ice,
and this allows precise dating of climatic events (Meese et
al., 1994). The air trapped in bubbles in the ice (-10% of
the ice volume) can be analyzed for CO2, nitrogen, and
methane, revealing changes in atmospheric chemistry (Sci-
ence, 1993); water from the ice can be studied for oxygen-
isotope composition, revealing changes in temperature.
Electrical signals also provide a record of conductivity
through the cores, which vary with the amount of wind-
blown dust. The amount of dust increases during cool-dry-
windy glacial intervals and reduces conductivity. The
record can also be tied to tree-ring chronologies (LaMarche
and Hirschboeck, 1984) and to volcanism (Zielinski et al.,
1994; but see Fiedel, 1995; Southon and Brown, 1995).
The GRIP and GISP2 cores, taken only about 28 km
apart, are similar down to -2700 m (100 Kya), which in-
cludes the present interglacial, the last glacial, and part of
the previous interglacial. Below that level, in the Sanga-
mon (North America) or Eemian (Europe) interglacial (ma-
rine isotope stage 5e), they differ (Grootes et al., 1993; Tay-
lor et al., 1993b); the differences have been attributed to
thrust faulting and distortion by ice flow over uneven ter-
ranes. These effects may extend above the last interglacial
boundary (Alley et al., 1995). Nonetheless, the records
from the upper levels of both cores show that in the period
from 40 to 8 Kya there were sudden changes of 5-10°C that
sometimes lasted less than 5 years, and in the past 8 Ky
changes of up to 10-12°C were recorded in a few decades
and lasted as little as 70 years. During the Younger Dryas
climates were dry, windy, and ~14±3°C colder than today,
while at the end of that interval near the Holocene bound-
ary (-10 Kya), it was wetter, less windy, and the average
temperature increase was +l°C/decade or steeper than the
most extreme forecasts for greenhouse warming during the
next 100 years. These brief oscillations on the order of 100
to a few thousands of years evident in the ice cores are
called Dansgaard-Oeschger (D-O) events (Bender et al.,
1994; Broecker and Denton, 1989; see Figs. 8.5, 8.6). Events
lasting longer than -2 Ky are also evident in the Antarctic
ice-core record, suggesting that they were of global extent
(Mayewski et al., 1996). The last D-O cycle was the
Younger Dryas (11-10 Kya), which was more evident in
Europe than in North America, and its end marked the be-
ginning of the Holocene. The principal transition took
place over a remarkably brief 40 years (Taylor et al., 1997).
Among the principal revelations of the Greenland ice-core
record is that in addition to shifts between glacial and in-
terglacial conditions at Milankovitch-time intervals, cli-
mates have switched rapidly between cold and warm
states over the past 100 Ky (and probably throughout the
Quaternary; Bond et al., 1997). Thus, additional climate-
controlling processes must be operating, and one likely
control is in the heat transport provided by ocean currents
(Birchfield and Broecker, 1990; Broecker, 1997; Broecker et
al., 1990; Lehman and Keigwin, 1992). The process is de-
scribed as the salinity-density ocean current model or the
thermohaline cell (the "conveyer belt").
Thermohaline Cell
The D-O events preserved in the Greenland ice cores clus-
ter into packets of temperature declines, and during the
most intensely cold phases a Heinrich event occurs fol-
lowed by a prominent warming. Heinrich events (after Har-
mut Heinrich, 1988) are spaced at -10 Ky (13-7 Ky) inter-
vals and are numbered H-l (top, beginning of Termination
1,14C age -14,500 years) through H-6 (-70 Kya; first phase
of the last glacial; see Marine Geology, 1996, for a collec-
tion of papers on Heinrich events). They are evident in ma-
rine sediments as zones of ice-rafted terrestrial debris,
which are associated with layers of depleted foraminifer
and other marine fossils, very cold temperatures, and low
surface salinity.2 Fluctuations on the time scales of D-O
and Heinrich events are not restricted to the high latitudes.
Grimm et al. (1993) noted vegetation trends in Florida that
correlate with the last five Heinrich events; moranes in
Chile and rock varnish (manganese-rich coatings on rock)
in the Great Basin appear to follow a similar pattern; cycles
of abundance of the marine alga Florisphem profunda fol-
lowing Heinrich events 1-5 occur in the equatorial Atlantic
(Mclntyre and Molfino, 1996); and D-O events are evident
in Antarctic ice cores (Vostok site, Dansgaard et al., 1993;
Raynaud et al., 1993). The cycles also match the NADW
variations to the south based on 13C/12C ratios in benthic
foraminifera.
Although these climatic changes appear global in ex-
tent, the mechanisms are presently unclear. Bond et al.
(1992,1993) found that layers of marine debris discovered
in the North Atlantic resulted from calving of numerous
icebergs as they expanded and neared the coasts at inter-
vals correlated with cold periods in the Greenland ice
cores. One source of the debris was the Laurentide ice
sheet via the Hudson Strait and the Gulf of St. Lawrence,

and similar layers were derived from Iceland. The layers
thin from west to east. There also were periods of increased
iceberg calving at intervals of 2-3 Ky corresponding to
longer D-O or mini-Heinrich events (Bond and Lotti, 1995).
The abundances in marine planktic foraminifera and the
layers of marine debris both varied in accordance with the
terrestrial ice-core records for the past 90 Ky, and all three
correlate well with the marine oxygen isotope record. This
provided the long-sought link between temperature and
the rapid changes observed in the ice cores and in the ma-
rine sediments. The question is, what causes the rapid tem-
perature changes?
One of the fundamental processes influencing world cli-
mates is the poleward transfer of heat from equatorial re-
gions to the high latitudes. This is accomplished by wind
and ocean currents (Fig. 2.6). As discussed earlier, warm
low-latitude waters flow northward via the Gulf Stream
into the North Atlantic Ocean. The lower ambient tempera-
ture cools the water and the flow of the westerlies across to
northern Europe increases evaporation, producing a cold,
saline-dense, heavy water that sinks and flows southward
(NADW). This pulls additional warm water from the south
in a conveyer belt system at an average flow rate of 20 mil-
lion m3/s, or -100 times that of the Amazon River. The pre-
vailing westerlies convey substantial heat toward Europe.
Clearly, any disruption of the conveyer belt would cool the
North Atlantic region in general and western Europe in
particular; the latter is confirmed by tree-ring analyses and
by fossil pollen and spore studies of Younger Dryas age
sediments (Bjorck et al., 1996). Rerouting of meltwaters
from the Mississippi River to the St. Lawrence drainage
with retreat of the ice margin has been proposed (Broecker
et al., 1989; but see de Vernal et al., 1996). Bond et al.
(1992) and Broecker (1994) suggest that the increased ice-
berg calving from the eastern North America Laurentide
and Iceland ice sheets and the melting of sea ice dilutes the
saline-dense North Atlantic water, preventing or reducing
the rate at which it sinks to form NADW. This in turn dis-
rupts thermohaline circulation (the "salt oscillator"). Mod-
els indicate that freshwater input of 0.06 x 106m3/S, about
one-fourth of the discharge of the Amazon River, could
shut down the conveyor (Manabe and Stouffer, 1995;
Rahmstorf, 1995).
The mechanisms causing the periodicity in dilution are
being explored through several models. One focuses on
glacial mechanics. When continental glaciers first begin to
form they are anchored (frozen) to the substrate and
thicken without appreciable spreading. As they thicken,
pressure on the base increases and the ice further serves as
an insulation layer that traps geothermal heat. Distortion
and melting of the base of the glacier causes the ice to flow
and calve icebergs (freshwater) into the ocean. The amounts
of water are estimated at half the volume of the present
Greenland ice sheet in a few hundred years. Upon thin-
ning, the glacier settles to the substrate, and the cycle re-
peats on a proposed -7 Ky cycle. However, the discovery
Cause and Effect 41
that the large Laurentide and small Iceland ice sheets both
discharged ice and meltwaters into the North Atlantic
Ocean on about the same schedule (Bond and Lotti, 1995)
suggests that glacial mechanics were a result and not a
cause of the climatic changes. Also, millennial-scale oscil-
lations occur in the Holocene when there were no large
northern ice sheets. The discovery that rapid climatic
changes correlated with the North Atlantic variability oc-
curred in the tropical Atlantic region (Hughen et al., 1996)
suggests that global forcing mechanisms are involved, pos-
sibly solar variability.
The Milankovitch variations provide one forcing mech-
anism for long-term climatic change. The ice-core data and
marine sedimentary record reveal the need for other mech-
anisms that explain sudden short-term reversals. Although
not fully understood, it is becoming increasingly likely
that such mechanisms operate by periodically disrupting
oceanic heat transfer to the North Atlantic region through
dilution by glacial meltwater (Johnson and Lauritzen,
1995; Keigwin et al., 1994; McManus et al., 1994; Paillard
and Labeyrie, 1994; Rahmstorf, 1994).
CO2 Concentrations
Much of the literature on atmospheric levels of CO2 and
other radiatively active or greenhouse gases [about 40, in-
cluding reactive trace gases such as ammonia (NH3), chlo-
rofluorocarbons (CFCs), methane (CHJ, nitrous oxide
(N2O), ozone (O3), and reduced sulfur (SO2) compounds]
deals with their effect on recent climate and human
economies (Crowley, 1991; Dacey et al., 1994; Dowlatabadi
and Morgan, 1993; Dunnette and O'Brien, 1992; Hansen et
al., 1991; Kareiva et al., 1993; Kaufmann and Stem, 1997;
Kelley and Wigley, 1992; Kiehl and Briegleb, 1993; Lutje-
harms and Valentine, 1991; Majumdar et al., 1992; Peters
and Lovejoy, 1992; Rind et al., 1992; Rubin et al., 1992;
Schlesinger and Ramankutty, 1992; Schneider, 1994; Sci-
ence, 1993, including cover illustration; Nature, 1992). For
example, it is known that the CO2 level rose from ~280
ppm in the preindustrial years of 1750-1800 to 315 in
1958, 338 in 1980, and 351 in 1988; it is expected to reach
600 ppm between 2030 and 2080 A.D. At present -6-8 gi-
gatons (Gt, billions of metric tons) of carbon are produced
annually through anthropogenic activities, which are
mainly industrial and automobile emissions and burning
of the rain forest. About one-half of this amount escapes to
the atmosphere; -2 Gt are taken up by the ocean; and much
of the remaining ~2 Gt is stored in the terrestrial biosphere
in soil, peat, and plant (e.g., the northern biosphere, Keel-
ing et al., 1996; deep-rooted grasses of the South American
savannas, Fisher et al., 1994) and animal biomass. Pro-
cesses and the net CO2 balance from among the various
components of the terrestrial realm are not well under-
stood, as indicated by the designates "the missing sink"
and "an enigmatic terrestrial reservoir" (Hanson, 1993;
Oechel et al., 1993; Siegenthaler and Sarmiento, 1993;
42 Late Cretaceous and Cenozoic History of North American Vegetation
Smith and Shugart, 1993). Measurements over a 55-day pe-
riod in the wet and dry seasons from a tropical rain forest
in Brazil showed the forest is a net absorber of CO2 (8.5 ±
2.0 mol/m2/year; Grace et al., 1995). In 1990 deforestation
in the low latitudes emitted 1.6 ± 0.4 petagrams (Pg =
1015/g = 1 Gt) of carbon, while forest expansion in the mid-
to high latitudes incorporated 0.7 ± 0.2 Pg into the bio-
mass, for a net input into the atmosphere of 0.9 ± 0.4 Pg
(Dixon et al., 1994). Natural changes in vegetation cover
(e.g., forest to desert) not only affect the albedo of the
Earth's surface and the amount of dust in the atmosphere,
but also the carrying capacity of CO2 by the biosphere. CO2
allows solar radiation of short wavelengths to penetrate the
atmosphere, but it retards the reradiation of longer wave
infrared heat energy. Modeling experiments show that a
doubling of CO2 concentration in the atmosphere raises the
mean annual surface temperature by 2-5°C (Geophysics
Study Committee, 1982; Wigley and Jones, 1981; 1.5-4.5°C,
Cess et al., 1993; Houghton et al., 1990, 1992). Increased
surface temperatures would have a significant impact on
agricultural patterns and require major readjustments in
local cultivation and management practices (Overpeck et
al., 1991).
The amount of CO2 in the atmosphere has also fluctu-
ated over geologic time (Berner, 1991; Crowley, 1993), and
within the past 15 Ky (Faure, 1990) to 160 Ky there is a
clear covariation with atmospheric temperature (Fig. 2.12).
The sources of these data for relatively recent times of
~100-300 Kya are many and varied. One is the measure-
Figure 2.12. CO2 and temperature records for the past
160 ka from Antarctic ice cores, and recent atmospheric
measurements. Reprinted from Hansen et al. (1993) with
the permission of the National Geographic Society.
ment of isotopic carbon contributed by the atmosphere and
by plant roots to form veins and nodules of soil carbonate.
Plants tend to take up more of the lighter 12C than the heav-
ier 13C, and the ratio can be used to distinguish between
the plant-derived component from root cell respiration and
the atmosphere-derived components in the paleosols. This
approach shows covariation between high versus low at-
mospheric CO2 concentration and warm versus cool tem-
peratures (Cerling, 1991).
A related technique involves distinguishing between
the 13C in paleosol carbonate contributed by C3 and C4
plants (Cerling et al., 1993; Morgan et al., 1994; see Na-
ture, 1994, for response and reply). In the photosynthesis
of C3 plants, CO2 is split and the carbon is taken up by
ribulose diphosphate, which splits into two molecules of
phosphoglyceric acid (Calvin cycle). The 13C composition
of C3 plants averages -26%. About 85% of flowering
plants follow the C3 pathway. In C4 plants the CO2 com-
bines with the 3-C phosphoenol pyruvate to form 4-C
malate or aspartic acid, which is transported to bundle
sheath cells (Hatch-Slack cycle). There CO2 is released
and used in the Calvin cycle reactions. The 13C composi-
tion of C4 plants averages -13%. The C4 pathway is de-
rived from C3, and approximately one half of the grasses
are C4 plants; these are mostly tropical and subtropical in
distribution. The C4 photosynthesis is favored under low
concentrations of CO2, while higher concentrations favor
the C3 pathway (Polley et al., 1993), and the isotopic car-
bon signature is preserved in plant debris from paleosols

and in ungulate tooth enamel. Beginning in the Middle
Miocene and especially by 7-6 Ma, plants with the C4
pathway began increasing in several parts of the world
(Cerling et al., 1997; Quade et al., 1989), suggesting that a
lower threshold of atmospheric CO2 had been reached
that, in turn, is consistent with evidence of progressively
lowering temperature and associated dryness during the
Neogene. An increase in atmospheric CO2 between 9 and 7
Kya (postglacial climatic optimum) is recorded in a shift
from C4-dominant grasses to C3-dominant shrubs on an
alluvial fan system in New Mexico (Cole and Monger,
1994; but see Boutton et al., 1994). Similar trends for the
past 100 Ma are preserved in carbon isotopes from marine
phytoplankton (Freeman and Hayes, 1992). Another tech-
nique being developed is the measurement of 18O in phos-
phate structures (e.g., fossil horse teeth 18.2-8.5 m.y. old
from Nebraska; Bryant et al., 1994). Results show a deple-
tion of 18O during this interval that is consistent with the
gradual trend toward lower temperatures and less evapo-
ration of the 18O.
Another source of information on ancient CO2 concen-
trations is from the 13C content of planktonic and benthic
foraminifera (Shackleton et al., 1983). These studies dem-
onstrate that the CO2 concentration of the atmosphere was
lower during the glacial ages than in the last interglacial,
and during the last glacial maximum at 18 Kya it was ~30%
lower than today. Atmospheric CO2 was ~220 ppm at 18
Kya, and 260-300 ppm during the interglacials. It is esti-
mated that the global content of carbon in the phytomass
varied between 968 Gt at glacial maxima (18 Kya) and 2319
Gt at present (Adams et al., 1990). The 13C/12C ratios in
mosses and sedges preserved in peat are also being used
(White et al., 1994). These analyses provide a resolution of
about a decade and show increases at 12,800 years B.P. (a
warming episode in the North Atlantic), 10 Kya (end of the
Younger Dryas cold period), and 4400 years B.P. (modern
interglacial climates).
The most detailed data are just now being published,
and they involve carbon isotopic analysis of air bubbles
of CO2 trapped in ice cores (Jouzel et al., 1993; Neftel et
al., 1982). These are from the Antarctica Vostok ice core
representing the past 500,000 years (Maier-Reimer et al.,
1990), the GRIP Camp Century and Dye 3 cores from
Greenland, and the GISP2 cores representing the past
150,000 years. These document a close relationship be-
tween CO2 concentration and polar temperatures for the
past 100,000 years.
Modeling experiments are also underway to determine
the effect of CO2 fluctuations on global climates in more
ancient times. Experiments for the Cretaceous at 100-65
Ma, using several alternative seafloor spreading rate for-
mulations, produce an atmospheric CO2 concentration
several times higher than at present (Berner, 1994; Berner
et al., 1983; near sevenfold, Berger and Spitzy, 1988), cor-
responding to the warm temperatures of that period. In
contrast, the cooling trend of the Middle to Late Cenozoic
Cause and Effect 43
involves a general decline in CO2 concentration to -2.5-
fold. A source of these data is air trapped in amber, in-
cluding material from the Dominican Republic (Late
Oligocene to Early Miocene), the Baltic (Oligocene to
Eocene), and Manitoba (Senonian, 75-95 Ma; Berner and
Landis, 1988; but see Technical Comments, Science,
1988).
It is becoming increasingly clear that changes in atmo-
spheric CO2 concentration have been a significant factor in
determining paleoclimates. The causes for these changes is
varied. As climates warm, land and water surfaces release
more CO2 as a positive feedback. This is one component of
the CO2-temperature relationship. Another derives from
geologic processes. A convenient way of envisioning these
processes is as a series of sinks (CO2 utilizing reactions)
and pumps (CO2 generating reactions) in which each rate
can be influenced by various geological and biological
events. The balance between them is important as evi-
denced by the fact that if today's sinks were just 10% larger
than the pumps, CO2 would disappear from the air-ocean
system in only 5 m.y.
Carbon dioxide is involved in the biological carbon
cycle through several processes. During the buildup of
complex organic molecules by plants and animals, it is
stored as fixed carbon (a sink); in decay and in the erosion
of buried organic material and kerogen, CO2 along with
ammonia and methane is produced (a pump). Although
these tend to balance out over long periods of time, on
shorter time scales there can be a significant lag period
with a resulting effect on climate. For example, the carbon
utilized in catabolic processes by organisms, which is
washed into the sea by erosion of organic-containing
coastal and shelf sediments and carried in by rivers, sinks
at the relatively rapid rate of -150 m/day. However, it
takes 100-1000 years before the CO2 released near the
ocean floor by dissolution is returned to the surface be-
cause of the slow circulation of the deep ocean. Periodi-
cally the release can be delayed even longer by changes in
surface temperatures, glacial meltwaters, land bridges,
salinity concentrations, and other factors. It is recognized
that the biological carbon cycle and factors that affect
ocean circulation and thermal stratification have an effect
on long-term climate through the release of CO2, but at
present these factors cannot be integrated and accurately
quantified.
In contrast, the effect of fluctuations in the geochemical
carbon cycle is clearer. In this cycle there can be significant
changes in the rate of production and consumption of CO2;
the recycling period is long, and it varies over time. A prin-
cipal cause for changes in atmospheric CO2 concentration
is variation in the rate of weathering and metamorphism of
silicate rocks (Berner, 1994; Berner et al., 1983). The pro-
cess is expressed by the following generalized reaction
[note the utilization of CO2 during weathering and later
burial storage (a sink) and its release during metamor-
phism (a pump)]:
44 Late Cretaceous and Cenozoic History of North American Vegetation
Any change in the rate of weathering or metamorphism
of silicate rocks will affect the atmospheric balance of CO2.
This is the basis for the contention by Raymo and Ruddi-
man (1992) and Molnar and England (1990) that uplift with
consequent increased rates of geochemical weathering of
the Tibetan Plateau, the Himalayas, and the Rocky Moun-
tains during the past 40 m.y. caused a drawdown of CO2
and was, therefore, a contributing factor to the resulting
cooler climates (see also, Nature, 1993; Partridge et al.,
1995).3
Another source of variation in CO2 concentration is the
process of degassing. CO2 is released at Benioff zones and
along midocean ridges by metamorphism and heating of
the water. In addition, CO2 can be generated from mineral
springs and from continental and oceanic flood basalts. A
rough estimate of the amount of CO2 emitted to the atmo-
sphere through degassing is -0.09 x 1015 g of C/year as CO2
(Arthur, 1982). Consequently, an increase in tectonic activ-
ity, as occurs during periods of major plate reorganization
(viz., in the Cretaceous), will release considerable quanti-
ties of CO2 to the atmosphere and contribute toward warm-
ing climates through greenhouse effect mechanisms. Sea-
floor spreading accounts for -80% of the world's volcanic
activity. Conversely, during times of tectonic quiescence,
less CO2 is released and there is a trend toward cooler cli-
mates as greater amounts of heat escape the troposphere.
Sea level can also affect CO2 concentration by influencing
the amounts of silicate and organic material available for
erosion (Chapter 3).
It is becoming clearer that CO2 has played other signifi-
cant but less direct and still poorly understood roles in pa-
leoclimatic change (Keir, 1992). Some studies suggest that
changes in CO2 concentration may amplify the 100 Ky Mi-
lankovitch cycle. In discussing the Milakovitch variations
it was noted that uncertainty still exists as to how the re-
distribution of relatively small amounts of heat from
changes in eccentricity could produce the dramatic cli-
matic effects of the past 1.6 m.y. The ice records for the
past 40 Ky from Camp Century, Greenland, and Byrd Sta-
tion and Dome C, Antarctica, that were studied by Neftel et
al. (1982) yielded CO2 concentrations of-180-240 ppm at
the last glacial maximum compared to 280 ppm for pre-
Industrial Revolution times. As the most recent glaciers
began retreating -16 Kya, CO2 concentration rapidly de-
clined. This pattern in the ice-core record is paralleled by
the marine isotope record from the eastern tropical Pacific
that was extended back to 340 Kya by Shackleton et al.
(1983). When changes in CO2 concentration were com-
pared with the Milankovitch variations, it was found that
there was first a change in the orbital cycle, followed by a
change in CO2 concentration and then a change in climate.
It was further documented that the greatest CO2 concentra-
tions matched the 100 Kya eccentricity period. The impli-
cation is that orbital variation may be a cause and CO2 fluc-
tuation an agent or amplifer of climatic change.
How these correlated changes in the Milankovitch vari-
ations, CO2 concentrations, and glacial cycles are brought
about is currently under study (Saltzman and Verbitsky,
1994). According to one scenario they involve the utiliza-
tion of CO2 by planktonic organisms to build their shells
and tissues. In the deep ocean the remains sink to great
depths where the carbonate dissolves and the organic mat-
ter is oxidized. The resulting carbon may be prevented
from returning immediately to the surface by the especially
steep temperature and salinity-density gradients that de-
velop at high latitudes during glacial times. The sea is the
primary carbon sink, and it presently stores an estimated
38 Gt of dissolved carbon (-65 times the amount in the at-
mosphere). Eventually the carbon does return to the ocean
surface at polar latitudes where more is released than can
be utilized by the plankton and it escapes back to the at-
mosphere as CO2. What causes fluctuations in the amount
and timing of CO2 circulating between the deep ocean, the
polar ocean surface, and the atmosphere has yet to be de-
termined. However, Milankovitch-induced warming trends
do raise surface water temperature and allow more CO2 to
escape. The full answer is still forthcoming, but changes in
CO2 concentration may amplify the eccentricity variant
and the mechanism likely involves chemical and physical
processes in the high latitude oceans (Oppo and Lehman,
1993).
PLATE TECTONICS
The patterns of climatic change induced by variations in
the orbital cycle are modified on a local to regional scale by
various consequences of plate tectonics (Gordon and Stein,
1992; Hill et al., 1992). These are described as telluric
changes in climate; that is, they are derived from alter-
ations in geography and topography that form the bound-
aries to the fluid spheres of water and atmosphere. The
most dramatic are the separation, transport, and collision
of continents over great distances. The separation of Aus-
tralia from Antarctica, beginning -50 Ma (early Eocene),
brought its south-temperate biota into contact with tropical
assemblages of southeast Asia as marked by Wallace's Line
between Java and Borneo and Lombok and the Celebes.

India separated from Africa by at least the Middle Creta-
ceous as is evident from oceanic basalts of that age between
Africa and Madascagar or India. The island continent
drifted northward from south-temperate, across equatorial-
tropical, into north-temperate zones until it collided with
island arcs located off south-central Asia in the Early
Eocene (50.3 Ma) but possibly as early as the Late Creta-
ceous (Beck et al., 1995). Thus, the time of early uplift of
the Himalayas and Tibetan Plateau resulting from the col-
lision is unsettled, but they were certainly substantially el-
evated by the Miocene (-20 Ma; Harrison et al., 1992).
Although the movement of India was geographically re-
moved a great distance from the North American conti-
nent, there were effects that illustrate the global interac-
tions between major changes in the physical environment,
world climate, and biotic history. Uplift of the Tibetan
Plateau altered atmospheric circulation, and the northward
drift of India diverted westward-flowing Pacific waters into
the counterclockwise subtropical gyre that presently char-
acterizes the southern Pacific Ocean. One consequence of
that diversion was to provide a moisture source for the
Antarctic continent that, along with the opening of the
Drake Passage that thermally isolated the continent, con-
tributed to the formation of glaciers there in the Oligocene.
Tasmania remained attached to Antarctica until -30 Ma, at
which time deep-water circulation was established around
Antarctica. The circum-Antarctica current transports -200
million m3 of water per second, which is probably the
largest volume of any ocean current. This, in turn, chilled
the southern oceans, affected ocean circulation, and re-
duced global sea levels. The Middle to Late Miocene dry-
ing trend evident from North American biotas was par-
tially due to colder ocean waters and reduced evaporation
resulting from the formation of glaciers on Antarctica.
North America experienced less longitudinal movement
across climatic zones during the Cenozoic. Movement is
estimated at an average rate of 2.2 cm/year during the past
10 m.y. (Anders, 1994), and the North American biota has
developed more under the influence of cycles of orbital-
induced climatic changes, augmented by other conse-
quences of plate tectonic activity and catastrophes. In par-
Cause and Effect 45
ticular, the role of plate tectonics in the evolution of North
American vegetation has been: to serve as the mechanism
for uplift, tilting, extension, compression, transpression,
and rotation of crustal blocks in the western part of the
continent (orogeny); volcanism in western North America;
fluctuating land barriers across Beringia to eastern Asia in
conjunction with sea-level changes; the generation of sim-
ilar connections and barriers across the North Atlantic to
western Europe; and the accretion of fossil-bearing exotic
or suspect terranes onto the western margin of the conti-
nent. Some of the results and chronologies of plate tectonic
activity relevant to the development of the North American
landscape are presented below and are summarized in
Table 2.2. A physiographic outline of North America, an
index of place names mentioned in the following text, and
a landform map (Thelin and Pike, 1991), are given in Figs.
2.13-2.15, respectively.
Orogeny
A valuable context for interpreting North American vegeta-
tional history and a major factor in modifying global and
regional climate is orogeny, the process of mountain build-
ing through uplift, folding, and thrusting of strata. In North
America the two principal erogenic systems are the Ap-
palachian Mountains and the western cordilleras includ-
ing the Rocky Mountains, Sierra Nevada, Cascade Moun-
tains, Coast Ranges, and other more local units.
The Appalachian Mountains extend 3000 km from cen-
tral Alabama to Newfoundland and include outliers such
as the Ouachita, Arbuckle, and Wichita Mountains in
Arkansas and Oklahoma, and a terminating western frag-
ment in the Marathon region of west Texas (King, 1977). To
the north, the system continues across the North Atlantic
into the British Isles and northwestern Europe as the Cale-
donian and Hercynian Mountains. The Appalachian Moun-
tains had their origin in ancient megasutures resulting
from collision of land masses that formed Pangaea and in
the breakup of Pangaea to form the central and North At-
lantic Ocean (Bally and Palmer, 1989; Sacks and Secor,
1990). The principal episode involving uplift of this sys-
46 Late Cretaceous and Cenozoic History of North American Vegetation

Table 2.2. Summary of Cretaceous and Tertiary events in development of North American landscape mentioned in text.

145 Ma (Early Cretaceous) Spreading evident in Labrador Sea

131 Ma (early Middle Cretaceous)
-112 Ma (Middle Cretaceous)
90 Ma (late Middle Cretaceous)
90-70 Ma (Late Cretaceous)
65 Ma (K-T boundary)
65-38 Ma (K-T boundary to Late Eocene)
33 Ma
30-29 Ma (Middle Oligocene)
27 Ma (early Late Oligocene)
26 Ma (middle Late Oligocene)
24 Ma (Oligo-Miocene)
23 Ma (Early Miocene)
20-18 Ma (Early Miocene)
18 Ma (Early Miocene)
17 Ma (Early Miocene)
15 Ma (Middle Miocene)
14 Ma (Middle Miocene)
10-5 Ma (Late Miocene)
6.5 Ma (Late Miocene)
3.4 Ma (Middle Pliocene)
Spreading evident in Davis Strait and southern Baffin Bay
Crustal separation between Labrador Shelf and Greenland
Ingression of Arctic sea into Baffin Bay rift
Mixing of Arctic-Atlantic waters marking formation of Davis Strait (70 Ma); Cape
Dyer (Baffin Isl.)-Disko Isl. (Greenland) separated by ~150 km (70 Ma); early for-
mation of Idaho Batholith (90-70 Ma); Blacktail, Snowcrest, Madison Ranges
(Montana; 70 Ma)
Asteroid impact
Volcanism increases in Davis Strait and north of Ellesmere Isl. (Thulean volcanism),
Cape Dyer-Disko Isl. separated by another 200 km (57-55 Ma); crustal separation
between Greenland and NW Europe; Rocky Mts. uplifted to -1 km (0.6 mi); frag-
mentation of the old Farallon plate to form the Juan de Fuca and other subplates
(55 Ma); beginning of Proto-Aleutian arc (55-50 Ma); beginning of Bitteroot
Dome (50-45 Ma); Uinta Range (45 Ma); formation of Great Basin topography
through extension and differential collapse (45 Ma north, 35 Ma south); begin-
ning of volcanism in the Cascades (44-38 Ma); major uplift of Aleutian arc (40 Ma)
Beginning of Tertiary volcanism in Sierra Nevada
Most Aleutian arc islands present (30 Ma); Pacific plate begins to impinge on west-
ern edge of North American continent
Subduction of Juan de Fuca plate activates Garibaldi-Pemberton volcanic belt (SW
Br. Columbia; to 7 Ma; then 2.5 Ma to present)
First Pacific ridge segments arrive off Baja California (28-26 Ma); San Andreas fault
structure develops (29 Ma), on-land activation (22 Ma), modern single-fault con-
figuration (12 Ma)
Uplift of Colorado Plateau ~3 km (2 mi)
Beginning uplift and volcanism in California Coast Ranges
Extension in Sonoran Desert (to -10 Ma); continued uplift of Rocky Mts.
Block faulting and differential uplift on Colorado Plateau (to 10 Ma), subsidence and
erosion to present
Cascade volcanic belt -100 km long; Columbia River basalts (17-6 Ma)
Aleutian arc essentially established in modern configuration
Major extension in Death Valley; Transverse Ranges (California; to Plio-Pleistocene);
subsidence Iceland-Faeroe Ridge with interchange of Norwegian-Greenland
Seas cold water with Atlantic Ocean warmer water
Rocky Mts. further uplifted; Cascade volcanic belt to present length and width (by
8-9 Ma); Alert Bay belt (Vancouver Isl.) formed (8-2.5 Ma)
Opening Gulf of California; Cascades and Coast Ranges begin to cast rainshadow
Latest and principal disruption of land through Beringia

tem was the Alleghem'an at 300-250 Ma (Middle Pennsyl-
vanian to Late Permian). The formation of the Appalachian
Mountains was completed by the end of the Paleozoic, and
they have since undergone 250 m.y. of erosion to produce
relatively low elevations and rounded domelike structures.
The highest peaks are Mt. Mitchell (2037 m, 6684 ft) and
Clingmans Dome (2025 m, 6642 ft) in the Blue Ridge
Mountains of North Carolina and Tennessee, respectively,
and Mt. Washington (1916 m, 6288 ft) in the White Moun-
tains of New Hampshire.
In contrast, the western cordilleras are younger (Late
Cretaceous to present), and the modern physiographic di-
versity is considerably greater. The highest peaks include
Mt. McKinley in the Alaska Range (6194 m, 20,320 ft), Mt.
Logan in the St. Elias Range of the southwest Yukon Terri-
tories (6050 m, 19,840 ft), Mt. Whitney in the Sierra
Nevada of California (4418 m, 14,494 ft), and several others
over 3400 m (11,000 ft) in elevation. The history of these
mountains has had a more direct influence on the develop-
ment of modern North American vegetation than that of
the older Appalachian Mountains.
In the Middle Jurassic (175 Ma) a foredeep developed at
the present site of the Rocky Mountains, followed by fore-
land subsidence of about 3 km between 84 and 66 Ma and
deformation and uplift of 1 km above sea level at 65 Ma
(Bird, 1988). Deformation through thrustfaulting intensi-
fied in the late Paleocene (-55 Ma) and waned during the
Eocene (-40 Ma). The period in the development of the
Rocky Mountains between -55 and 40 Ma constitutes the
latter part of the Laramide orogeny [Mid-Late Cretaceous
(Coniacian) to the Middle Eocene]. By the end of the
Eocene, erosion had produced a landscape similar to the
modern High Plains (Scott, 1975; but see below). Accord-
ing to the geologic data summarized by Ruddiman and
Kutzbach (1989, 1990) and Ruddiman et al. (1989), the
Rocky Mountains gained about half their present altitude
between 10 and 5 Ma. Regional uplift began in the south-
ern Rocky Mountains at -20 Ma (Middle Miocene) and in-
 Cause and Effect 47

1. TRAVERSE RANGES 11. BROOKS RANGE

2. COAST RANGES
3. GREAT VALLEY
4. SIERRA NEVADA
5. BASIN AND RANGE
6. COLORADO PLATEAU
7. KLAMATH MTS
8. OLYMPIC MTS
9. COLUMBIA PLATEAU
10. MACKENZIE MTS
12. INTERIOR MTS AND PLATEAUS
13. CASCADE MTS
14. IDAHO BATHOLITH
15. BITTEROOT MTS Figure 2.13. Physiographic
16. WIND RIVER MTS
17. BIGHORN MTS diagram of North America with
18.UINTA MTS
19. BLACK HILLS
reference to features mentioned
20. MISSISSIPPI EMBAYMENT in the text.

tensified at -12 Ma, and a particularly intense uplift oc-
curred between 7 and 4 Ma. The Rocky Mountains orogeny
is reflected by a change in sediments carried eastward onto
the High Plains by the Arkansas and Platte Rivers. At ~40
Ma these sediments were mostly clay and silt, changing to
sand at -20 Ma, then to pebbles and cobbles between 10
and 5 Ma. Downcutting of the Arkansas, Platte, and Col-
orado Rivers also increased during this time. Deposition of
eroded sediments and uplift along the flanks of the Rocky
Mountains raised the western plains to form the High
Plains. Downcutting by rivers indicates some uplift has oc-
curred within the past 5 Ma (Eaton, 1986). Uplift of the
Brooks Range along the North Slope of Alaska began in the
Berriasian (-145 Ma), continued into the Aptian (124 Ma),
and further intensified between the Albian (112 Ma) and
Turonian (90 Ma; Mull, 1979, 1985). The Idaho Batholith
formed between 90 and 70 Ma, the Blacktail-Snowcrest
and Madison Ranges in Montana were formed in the latest
Cretaceous, crustal shortening is evident in the British Co-
lumbia section of the southern Canadian Rockies in the Pa-
leocene, the Uinta Range formed in the middle Eocene
(~45 Ma), and the Bitteroot Dome was uplifted between 50
and 45 Ma.
The cause of the orogeny is complex and multifaceted,
but it involved a plate of Pacific oceanic lithosphere (the
Farallon plate, the ocean floor east of the mid-Pacific Rise;
Fig. 2.16) underthrusting the western edge of the North
American plate. The Farallon plate now has been mostly
subducted beneath North America; however, it fragmented
at -55 Ma into a series of smaller plates and subplates,
some of which are still impinging on the margins of the
continent (e.g., the Gorda and Juan de Fuca plates along the
coast of Washington and Oregon; Fig. 2.13; Clarke and
Carver, 1992; Savage and Lisowski, 1991). The rate of sub-
duction of the Farallon Plate during periods of intense tec-
tonism is estimated at -12 cm/year.
In sections from Wyoming the horizontal shortening of
strata caused by uplift was -5—10%. However, crustal
thickness here increased from 33 to 55 km, which was far
more than would have resulted from a shortening of 5-
10%. The extra material may be accounted for in a model
proposed by Bird (1988) in which the subduction track,
 1. WRANGEL ISL
2. SEWARD PENINSULA
3. KODIAK ISL
4. MEIGHEN ISL
5. SVERDRUP ISLS.
6. ELLESMERE ISL.
7. PRINCE PATRICK ISL
8. BANKS ISL
9. DEVON ISL.
10. DISKO ISL.
11. CAPE DYER
12. ICELAND-FAEROE RIDGE
13. MOUNT ST. HELENS

Figure 2.14. Index to place names mentioned in the text.

Figure 2.15, facing page. Landform map of conterminous United States (Thelin and Pike, 1991). U.S. Geological Survey
terminology:
Laurentian Upland: 1. Superior Upland. Atlantic Plain: 2. Continental shelf (not on map); 3. Coastal Plain, a. Embayed
section, b. Sea Island section, c. Floridian section, d. East Gulf Coastal Plain, e. Mississippi Alluvial Plain, f. West Gulf
Coastal Plain.Appalachian Highlands: 4. Piedmont province, a. Piedmont Upland, b. Piedmont Lowlands; 5. Blue
Ridge province, a. Northern section, b. Southern section; 6. Valley and Ridge province, a. Tennessee section, b. Middle
section, c. Hudson Valley; 7. St. Lawrence Valley, a. Champlain section, b. Northern section (not on map); 8. Appa-
lachian Plateaus, a. Mohawk section, b. Catskill section, c. Southern New York section, d. Allegheny section,
e. Kanawha section, £ Cumberland Plateau section, g. Cumberland Mountain section; 9. New England Province,
a. Seaboard Lowland section, b. New England Upland section, c. White Mountain section, d. Green Mountain section,
e. Taconic section, 10. Adirondack province. Interior Plains: 11. Interior Low Plateaus, a. Highland Rim section,
b. Lexington Plain, c. Nashville Basin; 12. Central Lowland, a. Eastern Lake section, b. Western Lake section, c. Wiscon-
sin Driftless section, d. Till Plains, e. Dissected Till Plains, f. Osage Plains; 13. Great Plains province, a. Missouri
Plateau (glaciated), b. Missouri Plateau (unglaciated), c. Black Hills, d. High Plains, e. Plains Border, f. Colorado Pied-
mont, g. Raton section, h. Pecos Valley, i. Edwards Plateau, k. Central Texas section. Interior Highlands: 14. Ozark
Plateaus, a. Springfield-Salem plateaus, b. Boston "Mountains"; 15. Ouachita province, a. Arkansas Valley, b. Ouachita
Mountains.Rocky Mountain System: 16. Southern Rocky Mountains, 17. Wyoming Basin, 18. Middle Rocky Mountains,
19. Northern Rocky Mountains. Intermontane Plateaus: 20. Columbia Plateau, a. Walla Walla Plateau, b. Blue Mountain
section, c. Payette section, d. Snake River Plain, e. Harney section; 21. Colorado Plateaus, a. High Plateaus of Utah,
b. Uinta Basin, c. Canyon Lands, d. Navajo section, e. Grand Canyon section, f. Datil section; 22. Basin and Range
province, a. Great Basin, b. Sonoran Desert, c. Salton Trough, d. Mexican Highland, e. Sacramento section. Pacific
Mountain System: 23. Cascade-Sierra Mountains, a. Northern Cascade Mountains, b. Middle Cascade Mountains,
c. Southern Cascade Mountains, d. Sierra Nevada; 24. Pacific Border province, a. Puget Trough, b. Olympic Mountains,
c. Oregon Coast Range, d. Klamath Moutains, e. California Trough, f. California Coast Ranges, g. Los Angeles Ranges;
25. Lower Californian province.

48
50 Late Cretaceous and Cenozoic History of North American Vegetation

from southwest to northeast, is simulated as nearly hori-

zontal, and crustal material is added (scraped) onto the
western edge of the developing Rocky Mountains. Finite-
element modeling techniques further reveal that the man-
tle layer of the North American lithosphere (at a depth of
~100 km) is stripped away in the subduction process and
displaced eastward as far as the Black Hills.
One effect of the uplift was the development of a zone of
extensional strain (a stretching and thinning of strata) im-
mediately west of the uplifting Rocky Mountains. The thin
crust, in combination with a reduced underlying mantle
lithosphere, resulted in a subsurface zone of heat and
structural weakness. Extension processes created sutures
for extrusion of volcanics and facilitated differential col-
lapse of the region (formation of grabens and half-grabens)
to create the topography of the Basin and Range Province.
The events primarily responsible for the modern configura-
tion of this province began in the Middle Eocene (-45 Ma)
in southern British Columbia and spread southeast to New
Mexico by the Oligocene (35 Ma; Bird, 1988). A major ex-
tension in Death Valley occurred during the last 14 m.y. In
the Sonoran Desert region, extension began in the Early
Miocene (20-18 Ma), immediately followed by rapid uplift
at a rate of -7.2 mm/year. The area has been relatively in-
active since 10 Ma. In general, the fast rate of regional ex-
tension was between 15 and 10 Ma.
The earliest dates for Tertiary volcanism in the Sierra
Nevada cluster around 33 Ma and mark the beginning of
that system. The Colorado Plateau, a tectonic block 45 km
in thickness, was uplifted -3 km beginning -24 Ma. Exten-
sion forces developed in the Early Miocene (20-18 Ma),
creating block faults, and portions were differentially up-
lifted by another kilometer with reference to the Basin and
Range Province floor between 18 and 10 Ma for a total up-
lift of -4 km. The present mean elevation is 2 km. The dif-
ference is likely due to erosion and subsidence resulting
from further extension in the Great Basin in Pliocene and
Quaternary times. Most of the net uplift has been since
-12-10 Ma.
With the Farallon plate subducted beneath North Amer-
ica, the present Pacific plate began to impinge on western
North America in the Middle to Late Oligocene (-29-30
Ma; Fig. 2.16). In terms of plate mechanisms, the contact
changed from oblique transpression (forces generated by
oblique rather than direct contact between plate margins;
Farallon plate) to a transform boundary (faults developing
through shearing as one plate slides past another; Pacific

Figure 2.16. Position of the principal plates impinging

on the North American continent during the four phases
of the Late Cretaceous and Cenozoic: (A) the present
(0-5), (B) 37 Ma (37-43), (C) 56 Ma (56-61), (D) 65 Ma
(65-71). Y, Yellowstone hot spot. Reprinted from
Winterer et al. (1989) with the permission of the
Geological Society of America.
 Cause and effect 51

A. Right-lateral displacements B. Left-lateral displacement

san andreas(SAF) 315 Garlock (GAF)
G 60
Grogan (GF 75 B
Big Pine (BPF) 14
Haywad-Rodgers Creak(h-RCF 43 Santa Ynez (SYF) 6+
Maacama (MAP) 20 M
Malibou Coast- 60-
1 Sunoi-Calaveras- 35 Santa Monica- 90
Franklin-Carneros (SCF) Raymond Hill-
San Gregorio-Hosgri (SG-HF) 115- Cucamonga
150 (MC-SM-CF)
Rinconada (RF) 60
San Gabriel (SGF) 60
East Santa Cruz Basin 120-
(ESCBF) 160
Elsinore (EF) 30
San Jacinto (SJF) 30

ODeath Valley (DVF)

Figure 2.1 7. Distribution and displacement (in meters) of major fault systems in California.
Reprinted from Oldow et al. (1989) with the permission of the Geological Society of America.

plate). The western edge of North America began to rise up Right-lateral slip now totals more than 315 km (Fig.
and over the east Pacific rise (the old Pacific—Farallon 2.17), and there are some 8000 tremors a year along the
spreading center, Fig. 2.16) by the Early Miocene (-17 Ma), fault. Uplift and volcanism in the California Coast Ranges
and this continues to the present. In the Los Angeles earth- began in the Oligocene (23 Ma), opening of the Gulf of Cal-
quake of January 17, 1994, and the strong aftershocks ifornia by sea-floor spreading began in the Miocene (Atwa-
through late March, the Sierra Nevada rose in places by ter, 1970), and transfer of Baja California to the Pacific plate
~380 cm (15 in.). The first Pacific ridge segments arrived began at 5-6 Ma. Baja California and areas of California
off northern Baja California 28-26 Ma (Winterer et al., west of the San Andreas fault are moving northward, rela-
1989). Among the consequences of the transform boundary tive to the rest of North America, at about 5 cm/year (Fig.
mechanics was the initial development of the 1000 km long 2.17). In 50 m.y. this will place the terrane off the coast of
San Andreas fault system at -29 Ma, and the first on-land Alaska. The Traverse Ranges of California rotated clock-
manifestations were at 22 Ma. The modern, single major- wise and rose by overthrusting and transpression in Mid-
fault configuration dates from -12 Ma, and the system at- dle Miocene to Plio-Pleistocene times. About two-thirds
tained its present form ~5 Ma. The San Andreas fault sys- of the elevation of the central Sierra Nevada was attained
tem consists of the fault plus about 40 other associated during the past 10 m.y. (Huber, 1981), and about two-thirds
ones that cover an area 200 km wide (Wallace, 1990). of the elevation of their west flank was attained since the
52 Late Cretaceous and Cenozoic History of North American Vegetation
Mio-Pliocene (5-6 Ma), with 1 km or more of uplift dur-
ing the past 3 m.y. The Sierra Nevada and the Cascade
Mountains began to cast a rainshadow to the east during
the latest Miocene and especially in the Pliocene. In the
Pliocene there was an increased differentiation of vegeta-
tion east and west of the Sierra Nevada, and at high eleva-
tions there was a change from temperate hardwoods and
evergreens to subalpine conifererous forest (Axelrod, 1962).
Tectonic deformation continued into Late Holocene and
modern times, as evidenced by earthquakes of 7.6-7.8
magnitude in the Humboldt Bay region of northern Califor-
nia in the past 1700 years [Clarke and Carver, 1992; see pa-
pers in Science (1992) regarding the Puget Sound region].
The central Coast Ranges of California were uplifted along
block faults during the past 2-3 m.y. (Christensen, 1965).
The volcanic arc that constitutes the Aleutian Islands
between western Alaska and Kamchatka developed in the
Eocene (55-50 Ma) through capture of an accretionary
wedge from the Cretaceous Kula plate. Subduction beneath
the North American plate was particularly active in the
Oligocene-Miocene and in the Pliocene-Quaternary. The
mountain massif including Mt. McKinley in the Alaska
Range underwent rapid uplift ~6 Ma (Fitzgerald et al.,
1993). At this time rivers that flowed south into the Gulf of
Alaska reversed to flow northward as at present. Volcanism
associated with principal uplift of the Tibetan Plateau
began at 13-14 Ma (Coleman and Hodges, 1995; Turner et
al., 1993); much of that elevation, which affected atmo-
spheric circulation patterns, was attained by -14 Ma (Cole-
man and Hodges, 1995) and at ~8 Ma (Harrison et al.,
1992). It should be noted that the Tibetan Plateau is the
largest topographic feature on Earth and has an average el-
evation of 5 km and a surface of over 5 million km2.
The uplift of the western North American mountains
and Asian plateaus had a significant and immensely com-
plex effect on the evolution of North American vegetation
through the creation of orographic rainfall and rainshadow
provinces, by altering atmospheric circulation, and by in-
creasing erosion of silicate rocks. The latter process draws
down CO2, allowing more heat to escape, and results in a
lowering of temperatures (see previous section on CO2). It
is ironic that even though an initial stimulus for the devel-
opment of large computers was to predict weather and as-
sess the impact of various factors on long-term climate, it is
still not possible to predict or precisely model the effect of
major changes in topography on global and regional cli-
mates. In fact, the permutations of such events are so com-
plex that weather predictions, climatic trends, El Ninos
(Jin et al., 1994; Tziperman et al., 1994), fluid dynamics (at-
mospheric and oceanic circulation patterns), and motion
of the inner planets are frequently used as examples of
nonlinear relationships that are dealt with in the revital-
ized field of mathematics called chaos theory (Allen et al.,
1993; Hanski et al., 1993; Lewin, 1992; Ruelle, 1991; Stew-
art, 1989; Waldrop, 1992). Chaotic or unforced fluctuations
(Fig. 2.18) are superimposed on those derived from Mi-
lankovitch cycles, continentality, orogeny, CO2 changes,
and other processes, and may be a particularly important
variable in ocean circulation and heat transport.
Nonetheless, it is generally recognized that uplift of
major mountain systems and plateaus affects climate, and
computer modeling is a convenient and powerful way to
investigate single factors involved in climate-orogeny in-
teractions (i.e., single-variable sensitivity tests). The CCM
has been used to simulate large-scale uplift of the Hi-
malayas and Tibetan Plateau and the Rocky Mountains and
Colorado Plateau (Kutzbach et al., 1989; Manabe and Broc-
coli, 1990; Manabe and Terpstra, 1974; Ruddiman and
Kutzbach, 1989, 1990; Ruddiman et al., 1989). In one sen-
sitivity test using the CCM at NCAR, the presence or ab-
sence of mountains and plateaus was altered (M, moun-
tains, simulating conditions at present; HM, half mountains,
10-3 Ma; NM, no mountains, with no exact paleophysio-
graphic equivalent, but closest to conditions assumed to
have prevailed in western North America prior to -40 Ma).
Other parameters were held constant [e.g., plate move-
ment, isthmian connections (Beringia, North Atlantic,
Panama, and the Drake Passage), CO2 concentrations, Arc-
tic sea ice, continentality, albedo effect of vegetation and
snow cover, and ocean temperature]. This allowed the ef-
fect of orogeny on atmospheric circulation and climate to
be isolated and more precisely defined. Such experiments
are run in perpetual January and July modes to determine
the effects on the extremes of annual climate.
In the models the general effect of mountain and plateau
uplift in Asia and western North America was a redirection
of the mean planetary waves in the northern hemisphere
troposphere at middle and high latitudes (Ruddiman and
Kutzbach, 1989). In the NM simulation (Fig. 2.19) a contin-
uous westerly circumpolar jet stream with relatively small
amplitude between troughs (low pressure) and ridges (high
pressure) in the Rossby regime is present at latitudes north
of 30° N. Patterns of annual precipitation are closely asso-
ciated with these storm tracks, and they are uniformly dis-
tributed across the continent (west to east). This most
closely approaches known conditions in western North
America in the Late Cretaceous and earliest Cenozoic prior
to extensive uplift. In the M experiment (Fig. 2.19) wind
tracks change from nearly zonal (viz., more uniform cli-
mates, both regionally and seasonally) to the meandering,
seasonal patterns of the present; the jet stream becomes
discontinuous; the amplitude of the troughs and ridges in-
creases; the mean positions of the troughs and ridges be-
come more stationary (viz., a large stationary trough devel-
ops to the lee of the Rocky Mountains with greater regional
differentiation of climate); wind flow, initially from the
west, becomes northwesterly along the western coast of the
United States by orographic deflection, which decreases
the moisture content of the avected air; air flow becomes
more northerly in the midwest; and a stronger southerly
component develops along the east coast, bringing warm,
moist summer winds into the region, as at present.
 Cause and Effect 53

Figure 2.18. Unforced (chaotic) global temperature variations in a 3 ka run of the

Global Climate Model (GCM) using the atmospheric composition of 1850 and a
mixed-layer ocean of 250-m maximum depth. Reprinted from Hansen et al.
(1993) with the permission of the National Geographic Society.

Climatic changes resulting from these alterations in at-
mospheric circulation are an increase in January precipita-
tion in eastern and southern North America and a decrease
in central North America (winter drying), an increase in
January precipitation over the southern and southeastern
United States and a decrease west of the Rocky Mountains,
drier summers along the American west coast (transition to
a Mediterranean climate) consistent with the gradual elim-
ination of summer wet vegetation, and drier winters in the
northern plains. Temperatures in the uplifted regions de-
crease by about 6°C (January cooling) because of rising ele-
vation and a stronger northerly flow of wind patterns;
colder winters and slightly cooler summers also develop in
eastern and central North America, which is consistent
with the elimination of much of the older (Eocene) sub-
tropical vegetation; and there is little change in the far
southeast, which is consistent with the persistence of ele-
ments of this vegetation into Late Cenozoic and modern
times. A summary of these changes is given in Table 2.3.
Thus, the presence of mountains produces a broad spec-
trum of climatic effects within the immediate region and in
distant parts of the world (Ruddiman, 1997). For example,
among the effects of the uplift of the western North Ameri-
can and central Asian mountains and plateaus, as pre-
dicted by the CCM, are a weakening and western shift of an
Icelandic low, where it merges with and reinforces a trough
located off the east coast of the United States (Ruddiman
and Kutzbach, 1989); a cooling of parts of the Arctic Ocean
and east-central North America because of stronger
northerly air flow; and a warming in Alaska because of a
stronger southerly air flow (see below).
Not all of the predictions of the CCM are borne out by
the geological record (Table 2.3), and refinements in the
model continue to be made (e.g., Sloan and Barren, 1990).
The present version calls for wetter summers and winters
in the southern Rocky Mountains and the High Plains, con-
trary to a wealth of paleobotanical information document-
ing a drying trend. This inconsistency is likely a result of
54 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 2.19. Computer simulation showing

the effect of uplift of the Tibetan Plateau and
Rocky Mountains on wind directions in July
for the North American continent. NM, no
mountains; M, mountains. Taken from
Ruddiman and Kutzbach (1990) and reprinted
with the permission of W. E Ruddiman and
the Royal Society of Edinburgh.

the relatively coarse grid constraints (Fig. El) that allow
the Gulf of Mexico to extend about 5° W of its actual posi-
tion, favoring transport of moisture to the High Plains and
southern Rocky Mountains (Ruddiman and Kutzbach,
1989). The model also calls for warmer winters in interior
Alaska, probably because it simulates broad plateaus more
precisely than it resolves large and relatively narrow
mountain systems like the Rocky Mountains into individ-
ual components having more local effects. Full-year inte-
grations will also improve the precision of the model.
In addition to effects on climate, orogeny affects the
course of development of regional vegetation in several
other ways. Increased topographic diversity provides a
greater number of habitats and increases the potential bi-
otic inventory and community types for a region. Uplift
and subsequent erosion creates new or disturbed habitats
that are important to the establishment and perpetuation of
hybrids. These interacting factors form a feedback system
within the model, because with increased biotic and habi-
tat diversity, the appearance and persistence of novel phe-
notypes accelerates, which further alters the composition
of the communities and increases the inventory.
The uplift of mountains is one form of barrier that can
divide the range of a species into geographically and, thus,

Table 2.3. Summary of simulated versus actual change in climate for North America based on CCM.
Region Simulated Change
W. coast NA Drier summers
Am. N. plains Drier winters (colder)
S. Rockies and S. plains Wetter summers and winters
E. NA Colder winters and summers
SE U.S. and Gulf Coast Wetter, little temp, change
Alaska Warmer winters, cooler summers
Adapted from Ruddiman and Kutzback (1989].
(+) indicates agreement between simulated and actual change; (-) indicates disagreement.
reproductively isolated populations, resulting in vicari-
ance speciation (Chapter 9). When substantial elevations
are achieved, the peaks constitute isolated climatic islands;
the arrangement of these peaks, in combination with the
dispersal potential of the propagules, provide pathways for
the migration of alpine, boreal, and temperate species dur-
ing periods of climatic change. The range expansion of
Pinus banksiana, Abies, and Picea into the southern Ap-
palachian Mountains during the Late Quaternary was
likely facilitated by increased target areas at high altitudes
during periods of cooling climate. The enhanced probabili-
ties afforded by topographic diversity for the establishment
of isolated populations further promotes speciation through
founder effect mechanisms, which are abrupt geographic
and reproductive isolation of segments of a population
through long-distance dispersal (Giddings et al., 1989). As
noted, mountains create orographic rainfall-rainshadow
systems and alter high-altitude atmospheric circulation,
leading to the development of different kinds of vegetation
on the windward and leeward sides.
The preceding scenario emphasizes mountain building
and plateau uplift as agents of climatic change that, in turn,
drive changes in vegetation through time. Molnar and Eng-
land (1990), Burbank (1992), and Small and Anderson
(1995) proposed an alternative model wherein climate is
an agent for late Cenozoic mountain and plateau uplift (see
also Caldeira et al., 1993; Raymo and Ruddiman, 1993;
Volk, 1993). They suggest that climate-induced increases
in mechanical erosion from Late Cenozoic cooling and
accompanying increases in precipitation and storminess
account for part of the uplift. Although an increase in the
elevation of highlands by erosion may at first seem para-
doxical, it is based on the principle of isostacy whereby
depressed surfaces rebound as overlying material is re-
moved. According to Raymo and Ruddiman (see also
Prell and Kutzbach, 1992) the two models may not be in-
compatible:
It is consistent with both hypotheses to argue that uplift
of Tibet (and possibly elsewhere) is a plausible first
cause of Cenozoic climatic changes (through circulation
changes and weathering), but that climatic change (in
particular, glacier activity) then caused additional ero-
sion, exhumation and isostatic uplift in other regions
. . . Thus, the arguments of Molnar and England can be
Cause and Effect 55
Data Vs. Model Observed Change
+ Drier summers
+ Drier winters (colder)
X Drier at low elev.; wetter (?) at high elev.
+ Colder winters (and summers?)
+ Little temp, change
X Colder, drier
viewed as a positive feedback mechanism, whereby
uplift-induced erosion initiates global cooling, which
then causes further glacial erosion and cooling world-
wide. (1992, p. 122)
In this combined version there would be initial uplift, re-
sulting in cooling via alteration of atmospheric circulation
and a decrease in CO2 concentration from erosion of sili-
cate rocks, some consequent impoverishment in vegeta-
tion, followed by increased erosion, more uplift, additional
climatic change, and further alteration in vegetation. Mol-
nar and England (1990) believe that considerably greater
uplift had been achieved in the Rocky Mountains earlier
than 10-7 Ma suggested by data summarized by Ruddiman
and Kutzbach (1989,1990) and Ruddiman et al. (1989) and
that the juvenile aspect of the terrane is due in large part to
recent climate-induced erosion followed by isostatic uplift
rather than to recent tectonic uplift. Current efforts at pa-
leoelevation analysis discussed in Chapter 6 suggest that
substantial elevations indeed may have been achieved in
parts of western North America by the Eocene-Oligocene.
In western Nevada elevations may have been higher than at
present because of compression forces associated with the
Laramide orogeny, followed by collapse from regional ex-
tension. This would allow the Late Cenozoic lowering of
temperature to be given a more prominent role in explain-
ing the biotic changes evident in the Middle Tertiary fossil
record. The relative importance of uplift and global tem-
perature change in influencing vegetational history is cur-
rently a subject of active discussion (Summerfield and
Kirkbride, 1992; see also reply by Molnar and England; Ax-
elrod, 1981). Although extremely complex, such hypothe-
ses are probably approaching the real-world intricacies of
erogenic - climatic -biotic interactions.
Disentangling the roles of orogeny and simultaneous
solar-induced change in climate is important in interpret-
ing vegetational history in western North America. In
many areas the older fossil assemblages, with prominent
warmth-requiring, broad-leaved, evergreen components,
are followed by deciduous vegetation characteristic of
cooler conditions. This involves uplift, providing cooler
high-altitudinal zones, and global cooling. Associated geo-
morphic evidence, such as increased downcutting by
streams, can be interpreted either as reflecting steeper gra-
dients (orogeny) or as a greater volume of water and trans-
56 Late Creataceous and Canozoic History of North American Vegetation
port load due to increased precipitation, and later due to
meltwater from developing alpine glaciers (climatic
change). If a real decrease in temperature of 6-9°C oc-
curred during the past 15 m.y., using a lapse rate of
~6°C/km, this would give a decrease in estimated uplift of
about 1500 m if the temperature decline was not taken into
account. As an example, MacGinitie (1953), using the mod-
ern analog method, estimated the paleoelevation of the
Late Eocene-Early Oligocene Florissant flora of central
Colorado at less than 915 m. Meyer (1992), using foliar
physiognomy, estimated MATs of lowland and upland fos-
sil floras, and using an assumed lapse rate of 5.9°C/km,
suggested a paleoelevation of 2450 m. The present eleva-
tion is -2600 m. Thus, one paleobotanical analysis sug-
gests uplift of about 1685 m since the Early Oligocene,
while more recent estimates suggest virtually no net uplift.
Computer simulations have clearly documented the
multifaceted effect of orogeny on atmospheric circulation
and subsequent surface climates. As expected, however,
the models also document that no single factor is sufficient
to account for the climate changes detected by 18O/16O ra-
tios in DSDP or OOP cores and evidenced by glaciations,
sea-level curves, and alterations in the Earth's biota. It is es-
timated that deep-water temperature in the early Eocene
was up to 12°C warmer than at present (Miller et al., 1987)
and that deep and surface waters were 15°C warmer in the
southern Pacific Ocean (Shackleton and Kennett, 1975).
High temperatures also characterized land surfaces in the
Early Eocene, then declined throughout the rest of the
Cenozoic. Orogeny can account for only some, albeit a
significant, part of this change: "[T]he amount of cooling
remains far short of that required to explain long-term
Cenozoic cooling. Additional factors, such as changes in
atmospheric composition, seem to be required to explain
global Cenozoic cooling" (Raymo and Ruddiman, 1992,
p. 118) (Table 2.2, Diagram above).
Volcanism
"[TJhere was a thick darkness in all the land of Egypt for
three days" (Exodus). [Eruption of Santorini (Thera),
Aegean Sea, second millenium B.C. (Rampino et al.,
1988)]
Another factor interacting with orbital-induced and oro-
genically influenced climatic trends is volcanism (Decker
and Decker, 1991; Fiocco et al., 1997; Wood and Kienle,
1990). Volcanic activity frequently occurs in association
with orogeny, is a direct consequence of plate tectonics,
and is generated through one of three mechanisms. Volca-
noes may form over hot spots, which are fissures in the
Earth's mantle through which deep-seated magma emerges
onto the surface as lava. This includes the type of volcanic
activity evident at Yellowstone National Park (Anders,
1994) and in the Hawaiian Islands. Another type results in
rift volcanoes that form where plates are diverging. These
volcanoes may be submerged or they may emerge along
midocean ridges as islands, such as Iceland. The third kind
of volcanism develops along subduction zones where the
edge of one plate dips under another along deep ocean
trenches bordering the active margins of continents. Rocks
at the leading edge of the subducting plate become molten
at depths usually between 100 and 150 km and rise to the

surface along fissures as arcs of volcanic activity called Be-
nioff zones (after Hugo Benioff, California Institute of Tech-
nology). The arcs may be offshore such as those of the
Antilles, Aleutians, Japanese, and Philippine Islands or in-
land such as those in western North and South America.
Volcanism in the Cascade Mountains began 44-38 Ma
(Middle Eocene) in connection with subduction of the Pa-
cific plate, and by 17 Ma (Early Miocene) a belt of volca-
noes 100 km wide extended along nearly the entire west-
ern margin of the continent. By 8-9 Ma (Late Miocene) the
arc had narrowed to its approximate present width and the
southern margin of the Cascade Mountains had retreated
northward to its present terminus at Mt. Lassen.
The Garibaldi-Pemberton volcanic belt in southwest
British Columbia was activated with the subduction of the
Juan de Fuca plate (a remnant of the old Farallon plate, Fig.
2.13) between 27 and 7 Ma, and the Alert Bay belt on Van-
couver Island formed between 8 and 2.5 Ma. Further vol-
canic activity was evident in the Garibaldi belt between 2.5
Ma and the present through subduction of the Juan de Fuca
and Explorer subplates.
To the north, the Colville foredeep formed on the north
shore of Alaska in the Early Cretaceous (~135-140 Ma),
which marks the beginning of the fold and thrust Brooks
Range. Uplift was strong into the Aptian (113 Ma) and be-
tween the Albian and the Turonian (113-85 Ma). The
proto-Aleutian arc began to develop 55-50 Ma as a result
of subduction of the Pacific plate beneath the North Amer-
ican plate. Between 80 and 45 Ma subduction was rapid,
estimated at 15-25 cm/year, and the volcanic arc formed as
a result of jamming of the subduction zone, with accretion
of Cretaceous strata between the trench and the continental
margin (Oldow et al., 1989). Major uplift occurred at 40 Ma
(Winterer et al., 1989), many of the islands had formed by
30 Ma, and the modern arc was essentially established
after -15 Ma. The highest elevations in the adjacent Alaska
Range (e.g., Mt. McKinley, 6194 m, 20,230 ft) were attained
within the past 5-6 m.y.
In recent times there has been increased opportunity to
study volcanic eruptions with advanced technology (e.g.,
the total ozone mapping spectrometer, or TOMS, Nimbus-7
satellite) and to more precisely quantify the effects on
global climate, particularly on temperature (Mass and Port-
man, 1989).
On March 20,1980, a seismograph operated by the Uni-
versity of Washington recorded an earthquake of Richter
Scale magnitude 4 on Mount St. Helens in the Cascade
Mountains of Washington. On March 25, 47 earthquakes
with magnitudes of 3 or more were registered within a 12-h
period; on May 18,1980, at 8:47 P.M., the first major volcanic
eruption occurred in the conterminous United States since
that of Lassen Peak in 1914-1917. The initial eruption was
followed by other explosions on May 25, June 12, July 22,
August 7, and October 16-18. The force of the explosions
was 1.7 x 1018 joules (J, 1 J= the energy of 1 amp of current
passed through a resistence of 1 ohm for 1 s). For compari-
Cause and Effect 57
son, a 1-megaton nuclear bomb generates 4.2 x 1015 J, so over
a 9-day period, Mount St. Helens produced energy equiva-
lent to 400 1-megaton bombs or the energy equivalent to
27,000 Hiroshima explosions at one every second for 9 h.
Mount St. Helens (Fig. 2.20), called Loowit (Lady of
Fire) by the Pacific Northwest Amerindians, erupted twice
about 13 Kya and 20 times in the last 4500 years, at an av-
erage interval of once every 225 years. Over the 9 days of
its most recent activity, -1 km3 of ash was discharged and
-2.7 km3 of volcanic rock was displaced. Mount St. Helens
is one of 15 active volcanoes in the Cascade Mountains
from Lassen Peak in California to Mt. Garibaldi in British
Columbia. Other volcanic eruptions equaling or surpassing
Mount St. Helens include:
Asama (Japan) 1783
Laki (Iceland) 1783
Mayon (Philippines) 1814
Tambora (Indonesia) 1815
Krakatoa (Indonesia) 1883
Mt. Augustine (Alaska) 1883
Tarawera (New Zealand) 1886
Mt. Pele, Soufriere (Antilles) 1902
Santa Maria (Guatemala) 1902
Ksudach (Kamchatka, Russia) 1907
Katmai (Alaska) 1912
Bezymianny (Kamchatka, Russia) 1956
Awu (Indonesia) 1966
Fuego (Philippines) 1974
Mt. Pinatubo (Philippines) 1991
Rabaul (Papua New Guinea) 1994
Some 400 active volcanoes are known to have erupted
around the Pacific ring of fire in historic times, and the
Earth presently has 37,000 km of subduction-zone volcanic
chains. There are over 260 volcanoes, volcanic complexes,
and volcanic fields in the western United States and
Canada (Wood and Kienle, 1990). Long Valley caldera in
California erupted 700 Kya, produced 500 km3 of ash, and
deposited 5 cm of ash across most of the central United
States. The Toba eruption of Sumatra 70 Kya was tens of
times larger than any historic eruption, and 2.2-4.4 billion
tons of sulfuric acid fell on the Earth over a 6-year period.
Mount Mazama exploded 7 Kya and formed Crater Lake in
Oregon.
In the 1980 Mount St. Helens activity, ash columns
reached an altitude of 18 km. Aerosol particles were car-
ried to Spokane, Washington, 430 km to the northwest,
where visibility was reduced to near darkness. Three days
after the eruption the ash had crossed the North American
continent, and aerosols still remain suspended in the
atmosphere.
In 1982 El Chichon (1260 m, 4130 ft), located in the
southern Mexican state of Chiapas near latitude 17° N,
erupted between March 28 and April 4 (Parker and Brown-
scombe, 1983; Rampino et al., 1988; Tilling et al., 1984).
Approximately 10-20 million metric tons (mt) of gaseous
sulfur dioxide and other aerosols were ejected and sent
58 Late Cretaceous and Cenozoic History of North American Vegetation
more than 25 km into the atmosphere, and ~0.3 km3 of py-
roclastic material was discharged (Hoffer et al., 1982). This
was nearly double the amount from Mount St. Helens.
Within a few weeks a narrow band of ejecta had encircled
the Earth, and within a year the stratosphere in most of the
northern hemisphere contained aerosols produced by the
eruption. The volcano was also active -600, 1250, and
1700 years ago and earlier at intervals of ~600 years. The
June 1991 eruption of Mt. Pinatubo in the Philippines
ejected 30 mt of aerosols into the atmosphere (Brasseur and
Granier, 1992; McCormick et al., 1995; Minnis et al., 1993)
and produced the greatest ash clouds observed during the
satellite era. Global cooling was estimated at 0.5°C for the
following 2-4 years. Estimates are 100 mt for Tambora,
and 50 mt for Krakatoa.
Ash tends to settle out within a few days; it is micro-
droplets of sulfuric acid, formed from ejected hydrogen
sulfide and sulfur dioxide, that mostly intercept and scatter
Figure 2.20. Eruption of Mount
St. Helens, May 18, 1980. Photo by
Austin Post; photograph number 17,
U.S.G.S. Photographic Library.
incoming insolation at ~25-km altitude (the Junge Layer;
Hofmann and Rosen, 1983; Kotra et al., 1983). The strato-
sphere warms because the aerosols are more effective at re-
flecting the shortwave solar radiation than the longwave
Earth-emitted radiation (Minnis et al., 1993). After the El
Chichon eruption, temperatures in the stratosphere over
the equatorial region rose by ~4°C. A warming of the upper
atmosphere through aerosol concentration reduces the
temperature cline, the strength of the Hadley Regime, and
the force of the westerly trade winds. The Earth's surface
(troposphere) cools because of reduced insolation, al-
though there is considerable local and seasonal variation
(Robock and Mao, 1992). The stratospheric aerosol cloud
of El Chichon was about 100 times more dense than the
ash-rich ejections of Mount St. Helens and will last for
years. The previous El Chichon eruption about 600 years
ago was several times the size of the latest one.
Continuing plate tectonic activity along the west coast

of North America is widely recognized through the devas-
tating and highly publicized earthquakes that periodically
affect the area. On April 25, 1992, a 7.1 earthquake oc-
curred at Cape Mendocino, California, along the subduc-
tion zone between the Gorda and North American plates
(Oppenheimer et al., 1993). Less well known is the fre-
quency of smaller earthquakes. In the 4-day period of Jan-
uary 18-22, 1991, the U.S. Geological Survey at Menlo
Park recorded 62 earthquakes in southern California regis-
tering 1 or more on the Richter Scale and 10 registering 2 or
more, which is one measurable earthquake every 90 min.
There is also evidence that residual spreading from the east
Pacific rise (the remnant of the mid-Pacific ridge between
the Pacific plate and the old Farallon plate) is still occur-
ring near the southern terminus of the Juan de Fuca plate.
At ~525 km west of the Oregon coast a string of 17 new
lava mounds, constituting 50 million m3 of new ocean
floor, appeared between 1981 and 1987. Associated with
the lava was the release of an estimated 100 million m3 of
water at 350°C. Such observations are important because
they provide insight into the nature of plate tectonically in-
duced events that shaped the evolution of western North
America during the Cenozoic.
Model simulations focusing on volcanism, holding
other factors constant, indicate that abrupt cooling of the
lower atmosphere occurs during the first 2-3 months after
a major eruption (Kelly and Sear, 1984) and that mean tem-
perature drops by ~0.1-0.3°C for 1-3 years. A lowering of
0.5°C occurred after the eruption of Krakatoa, and 1816
was known as "the year without a summer" after the 1815
eruption of Tambora. The drop after El Chichon was pre-
dicted to be ~0.4°C, but, because of the simultaneous
ENSO warming (Table 2.1), none was recorded. When the
ENSO effects are compensated for in the model, a drop of
several tenths of a degree results (Angell, 1988; Bradley,
1988; Mass and Portman, 1989). Models predicted a cool-
ing of ~0.5°C by late 1992 after the 1991 eruption of Mt.
Pinatubo. As of April 1992, a global cooling of ~0.4°C was
evident when adjusted for the 1992 ENSO event (Mc-
Cormick et al., 1995). The horizontal effects of an eruption
are detectable almost immediately because of rapid west-
east transport by directional winds. Effects to the north and
south are delayed a few months because the aerosol cloud
spreads more slowly by convection cell mechanisms.
There seems to be little or no effect on atmospheric pres-
sure or precipitation, even when large events are compos-
ited (Mass and Portman, 1989).
To better understand the effects of a few tenths of a
degree change in MAT on climate, it may be noted that
temperatures of the 1980s were generally high, 1990 was
then the warmest on record, and the winter of 1991-1992
was the mildest ever recorded. Media coverage showed
caravans of midwestern farmers carrying hay to drought-
stricken farmers in the southeastern United States. The
MAT for 1990 was 0.5°C above normal, and temperatures for
the 1980s were 0.34°C above previous decades. The year
Cause and Effect 59
1991 might have been even warmer were it not for the
eruption of Mt. Pinatubo.
The question relevant to Late Cretaceous and Cenozoic
vegetational history is whether aerosols resulting from vol-
canic eruptions have any long-term effect on climate. The
answer is still forthcoming because the sophisticated tech-
nology necessary to collect and analyze the data is new,
observations are constrained by the infrequency and un-
predictability of major eruptions, the response time of the
ocean—atmosphere system is 4—5 years after volcanic per-
tubation, and many other predictable and unpredictable
events are occurring simultaneously. As noted earlier, the
El Chichon eruption occurred during the anomalous ENSO
event of 1982, and the eruption of Mount Agung (Indone-
sia) in 1963 was followed by an ENSO event in 1963. Al-
though a cause and effect relationship has been suggested
(Handler, 1984), other analyses do not support such a hy-
pothesis (Mass and Portman, 1989). Nonetheless, when
ENSOs and eruptions occur simultaneously, it is more dif-
ficult to isolate the effects of volcanism alone on global cli-
mates. Such complex interactions continue in modern
times as shown by the August 1997 eruptions of Montser-
rat in the West Indies and predictions that the simultane-
ous El Nino would be the most intense since the 1982-
1983 event.
In addition to ENSO events and volcanism, the complex
nature of factors determining climates is further illustrated
by the fact that Late Cenozoic orogenies induced drying
trends in western North America, which reduced surface
vegetation and increased the amount of dust in the atmo-
sphere (Rea et al., 1985). Dust particles between 1 and 10
microns (um) in diameter may remain suspended indefi-
nitely and are distributed globally in the upper tropo-
sphere (Li et al., 1996; Tegen et al., 1996). An increase in
dust is recorded in oceanic sediments from the Pliocene
onward (Janecek, 1985; Leinen and Heath, 1981). As noted
by Ruddiman and Kutzbach (1989), the formation of local
or regional deserts through orogeny may load the atmo-
sphere with dust that produces a cooling effect on climate.
Data for soil erosion in modern times show that impressive
volumes can be involved. Sediment loss from forested re-
gions presently amounts to 1.7 t/ha/year, compared to 18
t/ha/year for croplands in the United States. The effect of
dust particles may be augmented by volcanism and molli-
fied by ENSO events. Carbon dioxide released by volcan-
ism tends to warm the atmosphere through a greenhouse
effect mechanism, while the simultaneous emission of other
aerosols lowers atmospheric temperatures.
Even with these complexities, there is evidence that
aerosols from volcanic activity influence global climates to
some degree by reducing insolation and altering atmo-
spheric circulation patterns (Zielinski et al., 1994). High
acid levels were recorded in ice layers from Greenland
(Hammer et al., 1980), corresponding to eruptions of
Krakatoa (1883), Tambora (1815), Laki (1783, Iceland), El-
dgja (10th century, Iceland), and Kekla (1104, Iceland). The
60 Late Cretaceous and Cenozoic History of North American Vegetation
highest and most sustained level occurred between 1350
and 1700 A.D., correlating with a period of cold climate and
glacial advance known as the Little Ice Age (Tilling et al.,
1984). Also, the eruption of Toba 70 Kya corresponds to the
early Wisconsin glaciation. However, the acidity spikes
may have been augmented, to some extent, by tropospheric
transport from volcanic activity in nearby Iceland (Ham-
mer, 1984; Rampino et al., 1988); temperature records from
the GRIPS2 ice core from Greenland now show that Toba
aerosols had no long-term effect on climate.
The pace of volcanic activity is a reflection of the rate of
subduction. For example, in Indonesia and Japan the an-
nual rate of subduction is presently ~6-7 cm/year and vol-
canic eruptions are frequent (-one/year), while in western
North America the slippage is 2-3 cm/year and eruptions
are less frequent. In comparison, convergence rates be-
tween the Juan de Fuca and North American plates de-
creased fivefold between 30 Ma and the present, and this
decrease correlates with a steady wane in volcanic inten-
sity (Oldow et al., 1989). Recall that the rate of subduction
of the Farallon plate has been as high as 12 cm/year and
that subduction of the Pacific plate along the Aleutian
Trench was 15-25 cm/year at 45 Ma.5 During times when
volcanic eruptions are more frequent and intense, they
probably contribute to a greater lowering of temperatures
for longer periods of time than the 3-year lowering of
0.3-0.4°C measured after the eruption of individual volca-
noes in recent times, perhaps affecting temperatures for
decades to a few hundred years (Mass and Portman, 1989).
Although volcanism is not presently regarded as a primary
forcing mechanism for long-term climatic change, depend-
ing on the timing, frequency, and coordination of volcanic
events, the effect may be to trigger change in an unstable
climate near a "bifurcation point" (Crowley and North,
1991) or to intensify or mollify trends controlled primarily
by orbital variations, major orogenies, CO2 concentration,
albedo changes, continentality, and other factors.
On a local level, accounting for the effects of volcanism
can be important in reconstructing vegetational history
and paleoenvironments based on fossil floras (Cross and
Taggart, 1982; Taggart and Cross, 1980). Thick deposits of
volcanic ash alter soil characteristics and reset the succes-
sional process (del Moral and Wood, 1993). In the absence
of adequate data on local or regional volcanism, the result-
ing changes in vegetation can be mistakenly attributed to
fluctuating climate.
In addition, volcanic eruptions have had consequences
related to other aspects of vegetational history. The mud
(lahars) and lava flows associated with volcanism dam
drainage systems, creating numerous lakes that serve as de-
positional basins for organic matter. In the case of Mount
St. Helens, mud flows along Toutle Creek crested 9 m (27
ft) above any previously recorded flood level. Between 17
and 6 Ma, 120-150 volcanic eruptions produced the Co-
lumbia River basalts, covering an area of about 200,000
km2 (somewhat larger than the state of Washington) with a
maximum thickness of 2.5 km (Hooper, 1982). The total
lava generated was over 700 billion m3 within the first few
days of the eruption. Similar flood basalts are known from
the older South African Karroo Series (175-195 Ma), the
South American Parana Series (133 Ma), and the Indian
Deccan Traps (65-69 Ma). Ash particles from these erup-
tions settled onto surfaces of the newly created lakes and
formed a fine-grained matrix for the preservation of fossils,
frequently with remarkable surface detail. The matrix may
also contain fossil pollen and spores and other plant and
animal microfossils. In addition, the ash charges the waters
with silicates that promote the growth of diatoms, resulting
in extensive deposits of diatomite, another fine-grained
sediment conducive to the preservation of plant and ani-
mal fossils. Many areas in western North America are char-
acterized by fossil-bearing shales of volcanic origin, such
as the Ruby Basin flora of the Late Eocene or Early Oligo-
cene age in southwestern Montana, or diatomite, such as
the Middle to Late Miocene Trout Creek flora of southeast-
ern Oregon. Frequently the deposits are associated with
thick sequences of lava (Figs. 2.21, 2.22) that provide mate-
rial for radiometric dating. Thus, an important conse-
quence of volcanism in western North America has been to
produce geologic conditions suitable for the preservation
and dating of one of the most extensive sequences of Ter-
tiary floras and faunas in the world (diagram p. 62).
Land Bridges
The role of land bridges in vegetational history is not sim-
ple. The most evident biological consequence is that at dif-
ferent times and to varying degrees they form a connecting
link, filter, or barrier to the migration of terrestrial organ-
isms and marine organisms. At the same time, however,
they alter climate by affecting ocean circulation. The con-
nections may be complete or partial, prevailing climates
may vary over time, and while links are being forged in one
place (e.g., the Isthmus of Panama), they can be broken
elsewhere (e.g., across the Bering Straits). During the Late
Mesozoic and Cenozoic the northern portions of North
America were connected to other continents via two land
bridges: one across the North Pacific to Asia and another
across the North Atlantic to Europe.
Beringia
In August of 1728 Russian explorer Vitus Bering sailed
through the strait between the Seward and Chukotka
Peninsulas, demonstrating that the New World and Old
World continents were indeed separate. Later soundings
showed, however, that water in the northeastern half of the
Bering Sea and portions of the Bering Strait and Chukchi
Seas was less than 100 fathoms (600 ft) deep and mostly
less than 100 ft. At the closest point North America is sep-
arated from Asia by only 80 km. The kinds of communities
are generally similar throughout the region, and paleonto-

Figure 2.21. Lava flow of the Columbia Plateau along highway 95 near the Idaho-Oregon border.
logical studies have documented a similarity among terres-
trial biotas on both sides of the strait for most of the Ter-
tiary. The area has long been recognized as an integrated bi-
otic and physiographic unit, and in 1937 Eric Hulten
applied the name Beringia to the region (Fig. 2.14).
From the standpoint of geology and plate tectonics, the
Cenozoic history of Beringia is complex. Numerous plates
and plate fragments (Pacific, Kula, Juan de Fuca) impinged
on the region, and exotic terranes consisting of rotated
blocks, volcanic arc segments, and forearc basins were
added to its margin. The fringing Aleutian Islands arose as
a volcanic arc above the subduction zone between the Pa-
cific and North American plates, beginning principally ~40
Ma; seamounts (guyots), originating over the Hawaiian Is-
lands hot spot, continue to be subducted into the Aleutian
trench. From the standpoint of physical geography the situ-
ation is simpler: the area was mostly above sea level for
most of the Tertiary. The disruption of land connections be-
tween North America and Asia dates from ~3 Ma (Repen-
ning and Brouwers, 1992), after which interchange was de-
termined primarily by Quaternary glaciations, sea-level
changes, and climate (Hopkins et al., 1982; West, 1996).
Continuous land was available between 60 Kya and 25 Kya
(oxygen isotope stage 3), and a sea-level fall of 121 m be-
tween 20 Kya and 18 Kya (late glacial; oxygen isotope stage
Cause and Effect 61
2) produced the maximum extent of the bridge during Late
Pleistocene time (-1500 km wide). After -11 Kya little ex-
change was possible (Hoffecker et al., 1993).
The North Atlantic
The history of land connections across the North Atlantic
is essentially the history of the breakup of northern Laura-
sia through the mechanism of plate tectonics (Dawes and
Kerr, 1982; Ziegler, 1988). Developing rift systems sepa-
rated Greenland from northwestern Europe through the
Denmark Strait and the Norwegian-Greenland Seas and
from mainland North America through the Labrador
Sea-Davis Strait-Baffin Bay region (Fig. 2.14). The separa-
tion involved north to south rifting in the North Atlantic
and adjacent Arctic Ocean in the early Mesozoic, subse-
quent development of a spreading center in the central At-
lantic, and eventual propagation of the northern terminus
of the mid-Atlantic Ridge system into the Arctic region.
In the Early Jurassic (-180 Ma) there was evidence of
crustal extension and differential subsidence of graben
structures on the Grand Banks and in the East Newfound-
land Basin as a result of rifting and wrench tectonics. In
general, wrench deformation (e.g., in northern Baffin Bay)
is a compensation for rifting in adjacent areas (e.g., in the
Figure 2.22. Ash deposits along highway 95 near the Idaho-Oregon border.

Labrador Sea). The Norwegian-Greenland Sea rift system
was also active. However, tectonic activity was consider-
ably less in both regions than in the preceding Triassic, and
the Mid-Jurassic actually represents a period of waning
influence for locally controlled rifting events. After the
Mid-Jurassic the center of plate tectonic activity shifted to
the central Atlantic as a continuation of the opening of the
Atlantic Ocean from the south (Ziegler, 1988). At this time
(180 Ma) actual crustal separation, as opposed to crustal
extension, was evident at the central mid-Atlantic Ridge;
by the Late Jurassic spreading was proceeding northward
at the rate of ~3.4 cm/year, decreasing to -2.3 cm/year in
the Early Cretaceous. By the Neocomian, spreading had
reached the southern part of the North Atlantic in the
Labrador Sea, as evidenced by dikes (tabular bodies of ig-
neous rock that cut across adjacent rocks) in southwestern
Greenland and diatremes (tubular volcanic vents through
flat, overlying rocks formed by explosive ejection of gas-
charged magmas) in southeastern Labrador. Crustal exten-
sion in the East Newfoundland Basin is estimated at be-
tween 100 and 200 km, while to the north it is much less.
By the Barremian (Early Cretaceous) rifting was also evi-
dent in the Davis Strait and in southern Baffin Bay. Ac-
cording to Ziegler (1988), during the Late Jurassic and
Early Cretaceous the Labrador-Baffin Bay Rift was extend-
ing into the southern Canadian Archipelago at a rate of ~8
cm/year. Extension in the Norwegian-Greenland Sea Rift
system also continued during the Late Jurassic and Early
Cretaceous.
In the Aptian-Albian, actual crustal separation devel-
oped between the Labrador Shelf and Greenland. North
Atlantic seafloor spreading extended rapidly into the
Labrador Sea, accompanied by crustal extension in Baffin
Bay induced by a counterclockwise rotation of Greenland
relative to North America. In the Turonian the Arctic Sea
invaded into the Baffin Bay Rift, and in the Maastrichtian
there is evidence of a mixing of Arctic and Atlantic waters
that marks the formation of the Davis Strait. Igneous activ-
ity in the Aptian—Albian through the Early Paleocene also
occurred along the north coast of Greenland as a result of
wrench movements between Svalbard and northeast Green-
land, which compensated for continued extension along
the Norwegian-Greenland Rift.
In the Late Paleocene to the earliest Eocene, volcanism
increased in the Davis Strait and north of Ellesmere Island.
This activity is known as the Thulean volcanism, and it
produced basalts estimated at volumes equivalent to those
of the Late Cretaceous Deccan basalts of India. An earlier
extension in the Cretaceous had separated Cape Dyer on
Baffin Island from Disko Island in west-central Greenland
by -150 km (Ziegler, 1988). During the latest Paleocene to
Early Oligocene the distance increased by another 200 km
through seafloor spreading. Spreading terminated in the
south Labrador Sea in the Early Oligocene. Crustal separa-
tion occurred in Baffin Bay during the Late Paleocene and
continued into the Early Oligocene. The period of Thulean
Cause and Effect 63
volcanism also represents the final stage in rifting before
initial crustal separation was achieved between Greenland
and northwest Europe in the earliest Eocene (-56 Ma;
chron 24; Eldholm et al., 1994). This brought to an end the
early Norwegian—Greenland Rift system that had affected
tectonics between Greenland and Europe for 275 Ma since
the Late Carboniferous.
Subsidence of the Iceland-Faeroe Ridge in the Middle
Miocene allowed interchange of cold waters from the Nor-
wegian—Greenland Seas with warmer waters from the At-
lantic Ocean. This event essentially established the present
configuration of the North Atlantic.
Several aspects of Late Cretaceous and Cenozoic paleo-
physiography of the North Atlantic region are important
from the standpoint of terrestrial biogeography. First, the
separation between North America and western Europe
progressed from south to north with propagation of the
mid-Atlantic Ridge. In a very general sense, therefore, two
land bridges can be recognized, particularly for the area be-
tween Greenland and Europe (McKenna, 1983). A southern
Thulean route, at 45-50° N paleolatitude, extended across
Greenland-Iceland into Scotland and afforded mostly con-
tinuous land surfaces through the end of the Cretaceous.
Although the mixing of Arctic and North Atlantic waters
through Baffin Bay and the Labrador Sea is first evident in
the Maastrichtian (-66 Ma), no major physical barriers ex-
isted for most terrestrial species until the Middle Eocene.
Fossil floras indicate that this southern portion of the North
Atlantic land bridge served as a migration route for tropical
and subtropical elements from the Late Cretaceous through
about the early Middle Eocene.
The northern portion of the land bridge through Sval-
bard (northern Scandinavia to northern Greenland), at pa-
leolatitude 55-60° N (to near 75° N; McKenna, 1972), is
known as the DeGeer route and it remained intact longer
and initial crustal separation was evident in the earliest
Eocene (-56 Ma). The DeGeer route probably provided
nearly continuous land surfaces until the Early Oligocene;
because of its northerly position, it was occupied in the
Late Cretaceous and Early Tertiary by more temperate and
temperate to subtropical transitional assemblages. After
that time the region was likely populated by progressively
cool to cold temperate vegetation, and glacial conditions
developed locally in the Miocene.
To some extent, reconstructing biogeographic events
through the North Atlantic land bridge is conditioned by
data from other regions. Recently it has been found that
genera such as Weigela (Caprifoliaceae, presently extinct in
North America) grew in the Arctic region during the Neo-
gene (Matthews and Ovenden, 1990), while present-day
North American genera such as Dulichium (Cyperaceae)
and Diervilla (Caprifoliaceae) are known from the fossil
record of Asia. Previously the North Atlantic connection
had to accommodate all such interchange, but recent stud-
ies document their presence in the Beringian region during
the Neogene. The Beringian pathway became increasingly
64 Late Cretaceous and Cenozoic History of North American Vegetation
more likely as the route of exchange of temperate elements
as rifting in the North Atlantic continued through the Neo-
gene. Thus, as a clearer picture emerges of Arctic vegeta-
tional history, it becomes unnecessary to overload the
North Atlantic land bridge with migrations that were phys-
ically, climatically, or temporally unlikely.
Second, the interchange of most terrestrial organisms is
not significantly affected during early phases of rifting in
which only narrow barriers exist. There is usually a lag
time between the inception of extension, crustal separa-
tion, and the appearance of distinct terrestrial biotic
provinces; and there have been numerous intervening land
surfaces across the North Atlantic Ocean throughout the
Late Cretaceous and Cenozoic. Many terrestrial organisms
continued to migrate through the region even after the ap-
pearance of channels, bays, straits, and narrow to moder-
ately broad seas.
Third, the structural separations were overprinted with
changes in sea level. Late Cretaceous sea levels are esti-
mated to have been as much as 300 m higher than at pres-
ent (Vail and Hardenbol, 1979), while during the maximum
of the Late Cenozoic glaciations they were from an esti-
mated 92 m to more than 100 m lower (Geophysics Study
Committee, 1990; Savin and Douglas, 1985). There were
other intervening fluctuations in land-sea relationships
that were caused by a variety of factors, several of which
are discussed by Ziegler (1988).
The interchange of terrestrial biotas during the Late Cre-
taceous and Cenozoic, like that through Beringia, was in-
fluenced but not determined solely by physiography and
sea levels. Changes in climate from the warm, equitable
conditions of the Late Cretaceous through the Early Eo-
cene, in the transitional period of the Middle Eocene and
Early Oligocene, to the glacial conditions of the Late Ceno-
zoic were also important in defining the timing and the
kinds of communities migrating across the North Atlantic
land bridge. The biotic elements and communities popu-
lating the region in the Late Cretaceous and Tertiary are de-
scribed in Chapters 5-8 (diagram p. 65).
Terranes
Another consequence of plate tectonics that is relevant to
North American vegetational history in the Late Creta-
ceous and Cenozoic is the terrane: sediments; portions or
entire volcanic island arcs, forearc basins (e.g., the Eel
River Basin of offshore California), and seamounts; frag-
ments of continents; and other suspect-exotic-allochtho-
nous (transported) materials that have been carried by
plate movement and accreted onto the active margins of
distant land masses (Dewey et al., 1991; Hashimoto and
Uyeda, 1983; Howell, 1989; Jones et al., 1983). One kind of
terrane is crustal blocks, which are fault-bound units with
a lithology, paleomagnetic configuration (remnant magnet-
ism), and/or fossil content that is different from adjacent
strata. An example is Wrangellia (from the Wrangell Moun-
tains 400 km east of Anchorage; Jones et al., 1977; Stone et
al., 1982). In the Triassic it was situated at or below the
equator and reached the vicinity of the paleo-Oregon-
British Columbia coast in the Middle to Late Cretaceous
(70-100 Ma). Pieces were later carried farther north through
nearly 24° of latitude by fault movement and are now
found on the Queen Charlotte Islands, and in the Wrangell
Mountains of southern Alaska. Present west-coastal Ore-
gon includes other fragments sutured on after Wrangellia
was in place. Support for the concept that parts of west-
coastal North America are from distant regions comes from
Vancouver Island. The remnant magnetism of mollusk-
bearing rocks indicates that in the Late Cretaceous these
rocks were located at ~P1 25±3°N, or off Baja California,
3500 km to the south (Ward et al., 1997).
Sediments may also be added to coastal margins in the
form of accretionary prisms when ocean-bottom material is
scraped onto continental margins at V-shaped subduction
zones. Accretionary prisms are found near Kodiak Island
and in the Gulf of Alaska. Thus, much of the west coast of
North America, from Baja California to Alaska, and extend-
ing inland up to 500 km, is a collage of some 100 exotic ter-
ranes added from Permian through Early Mesozoic time,
and especially between 200 (Early Jurassic) and 50 Ma
(Early Eocene; Churkin, 1983; Jones et al., 1983, fig. 1).
Terranes may be continental in size (e.g., India, which
includes a complex of smaller terranes) or comparatively
small, such as the individual basaltic seamounts that have
been periodically added to the Oregon coast. The distances
involved are from a few kilometers to thousands of kms for
terranes appressed to the California coast near San Fran-
cisco. The orientation may change during transport and
docking, as shown by the Channel Islands off southern Cal-
ifornia, which were rotated clockwise 70-90° during the
Neogene.
The implication of exotic terranes for vegetational his-
tory is apparent. If fossil-bearing units have been trans-
ported significant distances, paleoenvironmental recon-
structions are valid for the original locale of the terrane at
the time the fossils were being deposited, not for the even-
tual docking site. Similarly, biogeographic patterns (geo-
graphic relationships, pathways of migration, hypothe-
sized land bridges, and seaways) can be misconstructed if
the fossils were not part of the regional flora. A case in
point is the Miocene Mint Canyon flora of southern Cali-
fornia (Axelrod, 1940, 1986, 1987) that was transported
northward 200-300 km through some 2-3° of latitude (1°
of latitude = 110.68 km or 68.7 mi). Another example is the
Carmel flora (12-11 Ma) from the middle of the Monterey
Formation of California that was transported from the west
coast of Mexico (D. I. Axelrod, personal communication,
1996). During the Late Cretaceous the Salinia terrane (off
the coast of southern California) was located ~2600 km to
the south, and some Late Cretaceous volcanics north of San
Francisco (Point Arena) were probably displaced ~2900 km
between 90 and 48 Ma. The west coast of North America in

general, and Alaska in particular, includes terranes ac-
creted onto the continental margin. Recognition of this fact
is essential to the proper interpretation of North American
fossil floras along the west coast and for the accurate re-
construction of paleoenvironments (diagram p. 66).
CATASTROPHES
Catastrophes are devastating events in geologic time that
are unique or follow no apparent predictable pattern
(Ager, 1993; Clube, 1989; Huggett, 1990; Sharpton and
Ward, 1990). The effects may be global and influence the
subsequent course of evolution for various plant and ani-
mal groups via extinctions and new selective pressures,
including modification of world climates. The most prom-
inent example is the bolide (asteroid or comet) impact at
the end of the Cretaceous.6 Other effects are more regional
in scope. The late Wisconsin Missoula Floods devastated
extensive areas in the state of Washington, modifying
edaphic and landscape features and resetting successional
trends back to the initial colonization of barren rock. Fi-
nally, there are catastrophes occurring at present that are
part of the complex web of interacting factors influencing
global climates. One of the most tragic and potentially
dangerous to the welfare of human societies is the devas-
tation of the world's tropical forests and other natural veg-
Cause and Effect 65
etation. About 2-5% of the Earth's land area burns each
year, and this generates 5-25% of the radiative climatic
forcing from human-produced greenhouse gas emissions
(Levine, 1991; review by Overpeck, 1992). Plant transpira-
tion and evaporation are responsible for -65% of the total
precipitation falling on land or -70,000 km3 of water per
year. Response to forest clearing is rapid because water re-
sides in the atmosphere only -10 days before returning to
Earth. With nearly two-thirds of this plant formation al-
ready destroyed, the process constitutes a kind of mind-
less experiment on the climatic consequences of a major
alteration in the Earth's vegetation cover (Dunnette and
O'Brien, 1992).
Asteroid Impact
Paleontological records show that periods of increased ex-
tinction and ecosystem collapse followed by recovery and
appearance of new systems and lineages have occurred
several times in the past (Fig. 2.23; Belton et al., 1992,
1994). There was a notable die-off in the Late Devonian
known as the Frasnian-Famennian event and another at
the end of the Permian. One of the most prominent was the
terminal Cretaceous event (65 Ma) in which one-half of all
genera and 75% of all marine species were estimated to
have perished. Alvarez (1986) and Alvarez et al. (1980)
proposed that an asteroid collided with the Earth at the
end of the Cretaceous, sending large amounts of debris into
66 Late Cretaceous and Cenozoic History of North American Vegetation
the stratosphere (Covey et al., 1994) and creating darkness
and cooling that reduced the rate of photosynthesis. Other
consequences were shock waves, acid rain (Retallack,
1996), and global wildfires. The size of the asteroid was es-
timated at 10±4 km in diameter, the amount of ejecta at 60
times the object's mass, and the duration of the effects at
3-5 years. The force of the impact was estimated at 10,000
times the world's nuclear arsenal. An attractive feature of
the theory is that it provides a unique explanation for
widespread and sudden extinction in various plant and an-
imal groups that cannot be accounted for solely by changes
in climate [Crowley and North, 1988, 1991 (chapter 9);
Sheehan et al., 1991; see Marine Micropaleontology 29(2),
1996, for a series of papers on the El Kef blind test]. Even
the simultaneous occurrence of several reinforcing events
such as phases of the Milankovitch variations, ENSO
events, and volcanism (e.g., contemporaneous eruption of
the Late Cretaceous Deccan flood basalts) are inadequate,
as well as improbable, to account for the magnitude and
global extent of the extinctions. The oxygen isotope record
reflects no major deviation from the gradual temperature
decline for the 10-15 m.y. period bracketing the Cre-
taceous-Tertiary [K-T] boundary (Crowley and North,
1988).
An early basis for the theory was the discovery in 1978
of the platinum group or siderophile metal iridium (Ir) in
anomalous concentrations at the K-T boundary. Iridium is
depleted in the Earth's crust, but it is more abundant in
chondritic meteorites that cross the Earth's orbit. Amounts
of 20, 30, and 160 times the concentration in adjacent
strata have been found in K-T boundary clays. The initial
sites of these iridium anomalies were in Italy, New Zea-
land, and Denmark; but by 1985, 15 sites had been found
in such widespread localities as Spain, the South Atlantic,
the North Pacific, New Mexico, and Texas.
There were two principal objections to the original ver-
sion of the theory. First, not all plants and animals were
affected. Some animal groups, such as benthic foramini-
fera and some birds (Cooper and Penny, 1997) crossed the
boundary with little or no change, and others that disap-
peared near the K-T boundary were already in decline
prior to the impact. In particular, the rudists (extinct, shal-
low marine water bivalves with the habit of corals) disap-
peared about 1.5 m.y. before the event. Other bivalves,
such as the inoceramids, also show a gradual decline prior
to the end of the Cretaceous. Many terrestrial organisms,
including several plant groups, crossed the K-T boundary
without marked extinction. In contrast, planktic organ-
isms, the non-avian dinosaurs (near the top of the food
chain), and many ammonite species disappeared abruptly
(Marshall and Ward, 1996). A later impact of a meteorite
3.4 km in diameter at 50.8 Ma on offshore Nova Scotia re-
sulted in few extinctions (Jansa, 1993), and the conse-
quences of the Pliocene (2.15 Ma) Eltanin impact 1,500 km
southwest of Chile are still under study (Gersonde et al.,
1997). These observations introduce a selective component
to the mass extinction and require accommodation for
cooling climates and the gradual retreat of the seas from
Figure 2.23. Asteroid 243 Ida and its newly discovered moon photographed from NASA's Galileo spacecraft, August 28,
1993, from a distance of -10,870 km (6755 mi). The asteroid is -56 km (35 mi) long. It is in the main asteroid belt between
Mars and Jupiter and the 243rd discovered since the discovery of the first asteriod early in the 19th century. The photo-
graph was taken as the spacecraft flew past Ida on its way to Jupiter where it went into orbit in December 1995. Photo-
graph courtesy of the Jet Propulsion Laboratory, Pasadena, California. See also Belton et al. (1994).
68 Late Cretaceous and Cenozoic History of North American Vegetation
the continental interiors and margins. They demonstrate
that catastrophes, by definition, are superimposed on envi-
ronmental trends and evolutionary processes already in
progress and that no single event can be burdened with the
responsibility of accounting for the history and extinction
patterns of all groups. As noted by Clemens (quoted in
Morell, 1993; Clemens and Nelms, 1993), it is equally im-
portant for the theory to account for gradual and preimpact
extinctions and for survivals. These were inconsistent with
the proposed 3-5 year period of global near darkness orig-
inally proposed. Evidence presented at the Snowbird,
Utah, conferences in 1981 and 1988 (Sharpton and Ward,
1990) indicate that modern microfloras consume existing
food reserves within 10-100 days, a fact that allowed scal-
ing down of the dark period to 4-6 months. Extinctions at
the K-T boundary have been verified from DSDP cores at
-44% of all genera and -70% of the species in marine
plankton (compared to 14% of freshwater genera and 20%
of terrestrial genera).
A second problem was that no crater of the proper age
and size was known. Initially it was speculated that the
bolide landed in the ocean. Later it was suggested that the
Chicxulub (Mayan, tail of the devil) structure (180-km
diameter) buried -2 km beneath the Yucatan Peninsula
(40Ar/39Ar age of 64.98±0.05 m.y; Krogh et al, 1993; Swisher
et al., 1992), the Manson crater (36-km diameter) buried
under central Iowa (radiometric age originally reported at
-66 m.y.), and a site in the Pacific Ocean (2-km diameter,
DSDP 577, 32° 26' N and 157° 43' E, Shatsky Rise; Robin et
al., 1993) were multiple craters resulting from shattering as
the bolide entered the Earth's atmosphere (Hildebrand and
Boynton, 1990; Izett et al., 1991; Kring and Boynton, 1992;
Maurrasse and Sen, 1991; Sharpton et al., 1992). The diam-
eter of the Chicxulub crater was recently estimated at 300
km (Sharpton et al., 1993), but subsequent studies reaf-
firmed the 180-km value (Hildebrand et al., 1995). The di-
ameter is important because impact energy is about the
cube of crater size, so that the larger size would mean a
force 10 times greater than the smaller size. Chicxulub sam-
ples contain shocked quartz (stishovite, quartz grains with
striations that form under sudden intense pressure), and
samples from around the Gulf-Caribbean region contain
shocked quartz that splashed out from the impact site.
These are from Haiti, near Brazos, Texas, and Mimbral
north of Tampico, Mexico. The samples from the latter lo-
cality also contain ripple marks and debris from land veg-
etation that was washed into the ocean by an accompany-
ing tsunami (tidal wave) estimated at a kilometer or more
in height. Other evidence includes the presence of sidero-
phile (iron-rich) elements such as chromium, rhodium,
scandium, and titanium, and two amino acids that are rare
in Earth material but common in meteorites (Zhao and
Bada, 1989). Recently the isotopic composition of oxygen
and the impact glasses strontium and neodymium from
Chicxulub compared well with those from the Beloc For-
mation in Haiti (64.42±0.06 Ma; Dalrymple et al., 1993).
The isotopic chemistry of the Manson sediments is differ-
ent, which led Blum et al. (1993) to conclude that Chicxu-
lub is the primary source crater for the K-T catastrophy
and that the Manson melt rocks are unrelated to the event.
A new date on the Manson impact structure (73.8±0.3 Ma;
Izett et al., 1993), with splash (tsunami) debris in the co-
eval Crow Creek Member of the Pierre Shale of South
Dakota, supports this view.
The plausibility of an asteroid colliding with the Earth
in the past has been enhanced by reports of other collisions
(Grieve, 1996), as in the Late Eocene (Chesapeake Bay, Poag,
1997; Popigai crater in Siberia, Bottomley et al., 1997), re-
cent near misses, and near hits. The asteroid 1991 BA,
which was about 5-10 m in diameter, came within 170,000
km of the Earth (Scott et al., 1991). If it had hit, the velocity
of impact would have been 21.2 km/s and the explosive
force would have been -40 kilotons of TNT (3 times that of
the Hiroshima bomb). A near hit occurred over Tunguska,
Siberia, on June 30, 1908 (Chyba et al., 1993; Svetson,
1996), that felled trees over an area of 1000 km2, and over
one-third of the atmospheric ozone was destroyed by the
explosion. Asteroid particles were found embedded in tree
resin at the site. It is now estimated that the frequency of
impacts of near Earth objects (NEOs) 50 m in diameter is
about one per century. The break-up of comet Shoemaker 9
in the summer of 1994 provided spectacular documenta-
tion of a comet impact on Jupiter. These observations have
engendered some chilling proposals to avert future colli-
sions, including deflection by rocket-launched masses and
fragmentation by nuclear explosions (see discussions in
Ahrens and Harris, 1992; Sagan and Ostro, 1994).
Although various aspects of the theory are still being ar-
gued and refined, the consensus is that an impact did
occur at the end of the Cretaceous. In tracing the Late Cre-
taceous and Cenozoic history of North American terrestrial
vegetation and environments, it is thus necessary to moni-
tor the record for effects that may be related to this type of
catastrophy. Evergreen components were disappearing
from mesothermal vegetation of the midnorthern latitudes
at about this time, initiating a trend from evergreen to de-
ciduous vegetation. Temperatures were also declining from
Middle Cretaceous highs and may have reached a thresh-
old that favored development of deciduous vegetation, but
the trend could have been intensified by the bolide impact
(Wolfe, 1987; Chapter 5).
Missoula Floods
In eastern Washington and northern Idaho on the Columbia
Plateau, there is a vast area of erosional and depositional
land forms called the channeled scablands (Fig. 2.24). These
geomorphic features cover a region equivalent to the south-
eastern one-fourth of Washington or an area of-45,000 km2
(17,000 mi2). Between 1923 and 1969 J Harlen Bretz7 (e.g.,
Bretz et al., 1956) published a series of papers suggesting
that this unique landscape was a result of colossal floods

•^ .zfypttf
Figure 2.24. The scablands of eastern Washington resulting from the Missoula floods. Photograph courtesy of
Rachael Craig.
during the late Wisconsin glacial period (his Spokane
Flood). Bretz's ideas were initially rejected for a variety of
reasons. The event was unique and the scale incomprehen-
sible, there was no known source for the huge amounts of
water required, and no mechanism was proposed for the
sudden release of the water. The early 1900s was also a
time of rigid adherence to the gradualist principle, or uni-
formitarianism; processes acting today are those that have
acted in the past to produce the structures and events evi-
dent in the geological record. Random, unprecedented
events with no known cause evident from presently ob-
servable processes constituted catastrophism and were
summarily rejected. A sharp distinction between these
concepts is no longer recognized, and catastrophic events
are accepted as occasionally superimposed on uniformitar-
ian processes. As late as 1981 Raup noted that, "To con-
sider crises as a legitimate and important part of the forma-
tive process in ecology and evolution is new to many and
anathema to some." Derek Ager (1993) compared this view
of geologic history to the life of a soldier: long periods of
boredom and short periods of terror. Recent studies by
Baker (1973), Craig (1987), and Waitt (1985) provided sat-
isfactory explanations and mechanisms for the Missoula
floods, which are now documented as the most extensive
in recent Earth history. [See also Baker et al. 1993 regarding
a superflooding event in Siberia.]
Cause and Effect 69
The Purcell Trench lobe of the Cordilleran ice sheet ad-
vanced from the north during the Eraser glaciation (25-10
Kya) and abutted against the Bitteroot Range in northern
Idaho (Waitt, 1985, fig. 1). The ice lobe blocked the Clark
Fork River, creating glacial Lake Missoula to the east of the
ice dam. Flathead Lake in Montana is a remnant of glacial
Lake Missoula. The volume of the lake was -2500 km3 or
about Q times the amount of water in Lake Erie (458 km3).
When the lake reached a critical depth (-600 m), the ice
dam became buoyant and the seal against the mountain
was broken. The result was the release of vast amounts of
waters that scoured the landscape, carved elongated paral-
lel channels called coulees into the basalts of the Columbia
Plateau, and back-flooded up the Columbia River and its
tributaries. Ripple trains were formed with wavelengths of
150 m and amplitudes of 15 m; boulders 10 m in diameter
were carried more than 3.5 km; and discharge rates
reached 17 x 106 m 3 /s, which is more than 10 times the
combined flow of all the rivers of the world. These cata-
strophic floods are called jokulhlaups (Icelandic for out-
bursts) from similar events at Grimsvotn, Iceland. As the
water level fell, the ice settled back into place and the cycle
repeated. Waitt (1985) identified up to 60 of these hy-
draulic cycles between 15.3 and 12 Kya.
The Lake Missoula sediments are related stratigraphi-
cally to those of glacial Lake Bonneville to the south. Great
70 Late Cretaceous and Cenozoic History of North American Vegetation
Salt Lake in Utah is a remnant. Radiocarbon dates show
that the alluvial spillway damming the north end of Lake
Bonneville broke by head cutting between 15 and 14 Kya,
causing a sudden drop in the lake level of 115 m. The wa-
ters flowed through the Snake River Canyon to the Colum-
bia River, and near Lewiston, Idaho, the deposits are
merged with those of the Missoula floods. Evidence for cat-
aclysmic flooding is also found in Box Canyon along the
Big Lost River in east central Idaho (Rathburn, 1993). Dis-
charge is estimated at 60,000 m3/s, making it the third
largest known jb'kulhlaups.
The periodic removal of vegetation over a combined
area exceeding 45,000 km2, followed by relatively continu-
ous succession after 12 Kya, would be reflected in the com-
position of the regional pollen and spore rain onto adjacent
bog and lake surfaces. Therefore, high resolution pollen
and spore profiles derived from these sediments may re-
flect changes due to an exceptional and local environmen-
tal history superimposed upon regional climatic trends.
The Missoula floods further exemplify the intricacy and
dynamic nature of factors controlling environmental
change and influencing vegetational history. Embedded
within the rhythmites of back-flooded valleys are three
lenses of volcanic ash. Two are from eruptions of Mount St.
Helens -13 Kya, and one records the eruption of Mount
Mazama 7 Kya. Thus, the effects of volcanism discussed
earlier are added to the consequences of flooding,
long-term climatic trends resulting from variations in the
orbital cycles, fluctuations in the poleward transport of
heat, CO2 concentration changes, and other events. Collec-
tively, these constitute some of the complex and interacting
factors that determined late Wisconsin environments in the
region and influenced the course of development of the
plant formations.
Deforestation
[O]n Tuesday, July 22d [1494], he departed for Jamaica
. . . The sky, air, and climate were just the same as in
other places; every afternoon there was a rain squall that
lasted for about an hour. The admiral writes that he at-
tributes this to the great forests of that land; he knew
from experience that formerly this also occurred in the
Canary, Madeira, and Azore Islands, but since the re-
moval of forests that once covered those islands they do
not have so much mist and rain as before, (the biogra-
phy of Christopher Columbus, Colon, 1959)
From the end of the Cretaceous through the Middle
Eocene, warm-temperate to tropical vegetation extended
from equatorial regions to Beringia and the North Atlantic
land bridge. Subsequent climatic trends toward cooler and
seasonally dry climates have restricted the range of tropi-
cal forests in the northern hemisphere of the New World
today to below 17° N latitude near southern Veracruz, Mex-
ico. Many parts of the North American landscape once oc-
cupied by dense, evergreen forests now support deciduous
forests, grasslands, and deserts. In earlier discussions, cli-
mate, augmented and interacting with orogeny, volcanism,
various consequences of plate tectonic activity, and cata-
strophes have been emphasized as determining the compo-
sition and distribution of North American plant forma-
tions. It is worthwhile to note, however, that there is a
feedback circuit whereby vegetation influences climate
(Bonan et al., 1992; Foley et al, 1994; Kutzbach et al.,
1996). This is accomplished through at least three mecha-
nisms. First, various types of plant cover impart a different
albedo (pattern of heat reflection from surfaces; Berger,
1982, table 3.2). Albedo is -15 for forested land and -30 for
deserts (Table 3.1). Second, transpiration rates of water
into the atmosphere from leaf surfaces differ among vege-
tation types; and third, the type and extent of vegetation
cover affects the rate of erosion and, hence, dust and CO2
concentration.
It is not possible to specifically quantify the climatic ef-
fects of changes in plant cover from dense forest to a mo-
saic of open savanna, shrubland, grassland, and desert
vegetation. As with most large-scale multifactor events,
however, computer simulation models can aid in assessing
whether there is a potential effect, and if so, its theoretical
extent and how the variables interrelate. Shukla and Mintz
(1982) used the atmospheric GCM at Goddard Laboratory
for Atmospheric Sciences (GLAS) to simulate two ex-
tremes—an Earth completely covered with vegetation and
moist soils (wet-soil case) and one completely devoid of
vegetation (dry-soil case). Ocean-surface temperatures,
continental topography, and surface albedo were held at
current conditions.
In the wet-soil case precipitation in North America was
-3-6 mm/day, and in the dry-soil case it was reduced to 1
mm/day or less. (Eastern North America had more precipita-
tion from ocean water vapor.) The temperature simulations
show that regions north of -20° S latitude were 15-25°C
warmer in the dry-soil case. The conversion of high polar
landscapes from sparse tundra or bare soil to forest has a
modeled effect comparable to a doubling of CO2 concentra-
tion. From such data Shukla and Mintz (1982) conclude that
Calculations with a numerical model of the atmosphere
show that the global fields of rainfall, temperature, and
motion strongly depend on the land-surface evapotran-
spiration. This confirms the long-held idea that the sur-
face vegetation, which produces the evapotranspiration,
is an important factor in the earth's climate.
Using a more recent modified version of the CCM at
NCAR, McGuffie et al. (1995) also demonstrated a world-
wide effect of tropical deforestation on climate. From the
viewpoint of vegetational history this means that changes
in the composition and distribution of North American
plant formations through Cenozoic time are not simply a
response to climatic trends and alterations in the physical
environment. Rather, as environmental change alters the
vegetation cover, the alteration in turn influences climate,
Figure 2.25. Destruction of vegetation and subsequent
erosion north of Tehuantepec, Mexico.

Figure 2.26. Destruction of veg-

etation, Guatemala. Photograph
courtesy of Bruce Graham.
72 Late Cretaceous and Cenozoic History of North American Vegetation
either accelerating or countering precipitation and temper-
ature trends. The Food and Agriculture Organization (FAO,
United Nations), estimated 10 years ago that 70,000 km2 of
rain forest were being destroyed each year at 100 acres
every 3 min. The current rate is 100 acres/min, or an area
equal to the size of a football field every second. This is in
addition to 100,000 km2 that are drastically altered each
year, and other vegetation is being similarly impacted
throughout warm-temperate to tropical America (Figs.
2.25, 2.26). Salati et al. (1983; Salati and Vose, 1984) esti-
mate that half the current rainfall in the Amazon Basin is
associated with forest cover and that clear-cutting may
cause temperatures to rise by 5°C or more. Other aspects of
deforestation and climate are discussed by Crutzen and
Andreae (1990), Levine (1991), Sanford et al. (1985), Shukla
et al. (1990), and Woodwell et al. (1983).
The many factors influencing vegetational history dis-
cussed in this chapter are summarized in the diagram on
p. 72. In considering this illustration it is worth noting a
comment by Crowley and North:
The above studies indicate that it is unlikely that the
long-term evolution in past climates will be explained
by only one factor. One of the goals of paleoclimatology
is to integrate the effects of all of the above changes and
then compare the results with the long-term history
mapped by geologists. A substantial amount of addi-
tional work is required to reach this goal. (1990)
References
Adams, J. M., H. Faure, L. Faure-Denard, J. M. McGlade, and
E I. Woodward. 1990. Increases in terrestrial carbon
storage from the last glacial maximum to the present.
Nature 348: 711-714.
Ager, D. 1993. The New Catastrophism. Cambridge Univer-
sity Press, Cambridge, U.K.
Ahrens, T. J. and A. W. Harris. 1992. Deflection and frag-
mentation of near-Earth asteroids. Nature 360: 429-
433.
Allen, J. C., W. M. Schaffer, and D. Rosko. 1993. Chaos re-
duces species extinction by amplifying local popula-
tion noise. Nature 364: 229-232.
Alley, R. B. et al. (+ 5 authors). 1995. Comparison of deep
ice cores. Nature 373: 393-394.
Alvarez, L. W, W. Alvarez, E Asaro, and H. V Michel. 1980.
Extraterrestrial cause for the Cretaceous -Tertiary ex-
tinction. Science 208: 1095-1108.
Alvarez, W. 1986. Toward a theory of impact crises. EOS
(Trans. Am. Geophys. Union) 67: 649-658.
Anders, M. H. 1994. Constraints on North American plate
velocity from the Yellowstone hotspot deformation
field. Nature 369: 53-55.
Angell, J. K. 1988. Impact of El Nino on the delineation of
tropospheric cooling due to volcanic eruptions. J. Geo-
phys. Res. 93: 3697-3704.
Arthur, M. A. 1982. The carbon cycle—controls on atmo-
spheric CO2 and climate in the geologic past. In: Geo-
physics Study Committee, Climate in Earth History.
National Academy Press, Washington, D.C. pp. 55-67.

. and R. E. Garrison (eds.). 1986. Milankovitch cy-
cles through geologic time. Paleoceanography 1: 369-
586 [Special Section].
Atwater, T. 1970. Implications of plate tectonics for the
Cenozoic tectonic evolution of western North Amer-
ica. Geol. Soc. Am. Bull. 81: 3513-3536.
Axelrod, D. 1.1940. The Mint Canyon flora of southern Cal-
ifornia: a preliminary statement. Am. J. Sci. 238:
577-585.
. 1962. Post-Pliocene uplift of the Sierra Nevada,
California. Geol. Soc. Am. Bull. 73: 183-198.
. 1981. Altitudes of Tertiary forests estimated from
paleotemperature. In: Proc. of the Symposium on the
Qinghai-Xizang (Tibet) Plateau (Beijing, China). Sci-
ence Press, Beijing, and Gordon & Breach Science Pub-
lishers, New York. Vol. 1, pp. 131-137.
. 1986. Analysis of some palaeogeographic and
palaeoecologic problems of palaeobotany. Palaeobot-
anist 35: 115-129.
-. 1987. The late Oligocene Creede flora, Colorado.
Univ. Calif. Publ. Geol. Sci. 130:1-235.
Baker, V. R. 1973. Paleohydrology and Sedimentology of
Lake Missoula Flooding in Eastern Washington. Geo-
logical Society of America, Boulder, CO. Special Paper
144.
, G. Benito, and A. N. Rudoy. 1993. Paleohydrology
of late Pleistocene superflooding, Altay Mountains,
Siberia. Science 259: 348-350.
Bally, A. W. and A. R. Palmer (eds.). 1989. The Geology of
North America; An Overview. DNAG (Decade of North
American Geology). Geological Society of America,
Boulder, CO. Vol. A.
Barbour, M. G. and W. D. Billings (eds.). 1988. North Amer-
ican Terrestrial Vegetation. Cambridge University
Press, Cambridge, U.K.
and N. L. Christensen. 1993. Vegetation. In: Flora of
North America Editorial Committee (eds.), Flora of
North America North of Mexico. Oxford University
Press, Oxford, U.K. Vol. 1, pp. 97-131.
Bard, E., E Rostek, and C. Sonzogni. 1997. Interhemi-
spheric synchrony of the last deglaciation inferred
from alkenone palaeothermometry. Nature 385: 707-
710.
Barry, R. G. and R. J. Chorley. 1992. Atmosphere, Weather,
and Climate. Routledge, London. 6th edition.
Beck, J. W, J. Recy, E Taylor, R. L. Edwards, and G. Cabioch.
1997. Abrupt changes in Early Holocene tropical sea
surface temperature derived from coral records. Na-
ture 385: 705-707.
Beck, R. A. et al. (+13 authors). 1995. Stratigraphic evi-
dence for an early collision between northwest India
and Asia. Nature 373: 55-58.
Belton, M. J. S. et al. (+ 19 authors). 1992. Galileo en-
counter with 951 Gaspra: first pictures of an asteroid.
Science 257: 1647-1652.
et al. (+ 19 authors). 1994. First images of asteroid
243 Ida. Science 265: 1543-1547.
Bender, M. et al. (+ 6 authors). 1994. Climate correlations
between Greenland and Antarctica during the past
100,000 years. Nature 372: 663-666.
Berger, A. L. 1980. The Milankovitch astronomical theory
of paleoclimates: a modern review. Vistas Astron. 24:
103-122.
, J. Imbrie, J. Hays, G. Kukla, and B. Saltzman (eds.).
Cause and Effect 73
1984. Milankovitch and Climate. NATO ASI Series C:
Mathematical and Physical Sciences. D. Reidel Publ.
Co., Dordrecht. Vol. 126, Parts 1 and 2.
-, M. E Loutre, and J. Laskar. 1992. Stability of the as-
tronomical frequencies over the Earth's history for pa-
leoclimate studies. Science 255: 560-566.
Berger, W. H. 1982. Climate steps in ocean history—lessons
from the Pleistocene. In: Geophysics Study Committee
(eds.), Climate in Earth History. National Academy
Press, Washington, D.C. pp. 43-54.
and A. Spitzy. 1988. History of atmospheric CO2:
constraints from the deep-sea record. Paleoceanogra-
phy 3: 401-411.
Berggren, W. A. 1982. Role of ocean gateways in climatic
change. In: Geophysics Study Committee (eds.), Cli-
mate in Earth History. National Academy Press, Wash-
ington, D.C. pp. 118-125.
Berner, R. A. 1991. A model for atmospheric CO2 over
Phanerozoic time. Am. J. Sci. 291: 339-376.
. 1994. 3GEOCARB II: a revised model of atmo-
spheric CO2 over Phanerozoic time. Am. J. Sci. 294:
56-91.
and G. P. Landis. 1988. Gas bubbles in fossil amber
as possible indicators of the major gas composition of
ancient air. Science 239:1406-1409.
-, A. C. Lasaga, and R. M. Garrels. 1983. The carbon-
ate-silicate geochemical cycle and its effect on atmo-
spheric carbon dioxide over the past 100 million
years. Am. J. Sci. 283: 641-683.
Bevis, M. et al. (+ 10 authors). 1995. Geodetic observations
of very rapid convergence and back-arc extension at
the Tonga Arc. Nature 374: 249-251.
Birchfield, G. E. and W. S. Broecker. 1990. A salt oscillator
in the glacial Atlantic? 2. A "scale analysis" model. Pa-
leoceanography 5: 835-843.
Bird, J. 1991. Supercomputer voyages to the southern seas.
Science 254: 656-657.
Bird, P. 1988. Formation of the Rocky Mountains, western
United States: a continuum computer model. Science
239: 1501-1507.
Bjork, S. et al. (+ 10 authors). 1996. Synchronized terres-
trial-atmospheric deglacial records around the North
Atlantic. Science 274: 1155-1160.
Blum, J. D. et al. (+ 6 authors). 1993. Isotopic comparison of
K/T boundary impact glass with melt rock from the
Chicxulub and Manson impact structures. Nature 364:
325-327.
Bonan, G. B., D. Pollard, and S. L. Thompson. 1992. Effects
of boreal forest vegetation on global climate. Nature
359: 716-718.
Bond, G. C. and R. Lotti. 1995. Iceberg discharges into the
North Atlantic on millennial time scales during the
last glaciation. Science 267:1005-1010.
et al. (+ 13 authors). 1992. Evidence for massive dis-
charges of icebergs into the North Atlantic Ocean dur-
ing the last glacial period. Nature 360: 245-249.
et al. (+ 6 authors). 1993. Correlations between cli-
mate records from North Atlantic sediments and
Greenland ice. Nature 365:143-147.
et al. (+ 9 authors). 1997. A pervasive millennial-
scale cycle in North Atlantic Holocene and glacial cli-
mates. Science 278: 1257-1266.
Bottomley, R., R. Grieve, D. York, and V. Masaitis. 1997.
The age of the Popigai impact event and its relation to
74 Late Cretaceous and Cenozoic History of North American Vegetation
events at the Eocene/Oligocene boundary. Nature 388:
365-368.
Boutton, T. W, S. R. Archer, and L. C. Nordt. 1994. Scien-
tific correspondence: Climate, CO2 and plant abun-
dance. Nature 372: 625-626.
Bradley, R. S. 1988. The explosive volcanic eruption signal
in northern hemisphere continental temperature
records. Climatic Change 12: 221-243.
Brass, G. W. et al. (+ 6 authors). 1982. Ocean circulation,
plate tectonics, and climate. In: Geophysics Study
Committee (eds.), Climate in Earth History. National
Academy Press, Washington, D.C. pp. 83-89.
Brasseur, G. and C. Granier. 1992. Mount Pinatubo aerosols,
chloroflurocarbons, and ozone depletion. Science 257:
1239-1242.
Brenchley, R J. (ed.). 1984. Fossils and Climate. Wiley-Inter-
science, New York.
Bretz, J H., H. T. U. Smith, and G. E Neff. 1956. Channeled
scabland of Washington: new data and interpretations.
Bull. Geol. Soc. Am. 67: 957-1049.
Broecker, W. S. 1994. Massive iceberg discharges as triggers
for global climate change. Nature 372: 421-424.
. 1997. Thermohaline circulation, the Achilles Heel
of our climate system: will man-made CO2 upset the
current balance? Science 278: 1582-1588.
, G. Bond, M. Klas, G. Bonani, and W. Wolfli. 1990. A
salt oscillator in the glacial Atlantic? I. The concept.
Paleoceanography 5: 469-477.
and G. H. Denton. 1989. The role of ocean-atmo-
sphere reorganizations in glacial cycles. Geochim. et
Cosmochim. Acta 53: 2465-2501.
and J. van Donk. 1970. Insolation changes, ice vol-
umes, and the O18 record in deep-sea cores. Rev. Geo-
phys. Space Phys. 8: 169-198.
et al. (+ 5 authors). 1968. Milankovitch hypothesis
supported by precise dating of coral reefs and
deep-sea sediments. Science 159: 297-300.
et al. (+ 6 authors). 1988. The chronology of the last
deglaciation: implications to the cause of the Younger
Dry as event. Paleoceanography 3: 1-19.
et al. (+ 6 authors). 1989. Routing of meltwater from
the Laurentide ice sheet during the Younger Dryas
cold episode. Nature 341: 318-321.
Brownlee, D. E. 1995. A driver of glaciation cycles? Nature
378: 558.
Bryant, J. D., B. Luz, and R N. Froelich. 1994. Oxygen iso-
topic composition of fossil horse tooth phosphate as a
record of continental paleoclimate. Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 107: 303-316.
Burbank, D. W. 1992. Causes of recent Himalayan uplift de-
duced from depositional patterns in the Ganges Basin.
Nature 357: 680-683.
Caldeira, K., M. A. Arthur, R. A. Berner, and A. C. Lasaga.
1993. Scientific correspondence: Cooling in the Late
Cenozoic. Nature 361: 123-124.
Cerling, T. E. 1991. Carbon dioxide in the atmosphere: evi-
dence from Cenozoic and Mesozoic paleosols. Am. J.
Sci. 291: 377-400.
, Y. Wang, and J. Quade. 1993. Expansion of C4
ecosystems as an indicator of global ecological change
in the late Miocene. Nature 361: 344-345.
Cerling, T. E. et al. (+ 6 authors). 1997. Global vegetation
change through the Miocene/Pliocene boundary. Na-
ture 389: 153-158.
Cess, R. D. et al. (+ 29 authors). 1993. Uncertainties in car-
bon dioxide radiative forcing in atmospheric general
circulation models. Science 262:1252-1255.
et al. (+ 19 authors). 1995. Absorption of solar radi-
ation by clouds: observations versus models. Science
267:496-499.
Chahine, M. T. 1992. The hydrological cycle and its influ-
ence on climate. Nature 359: 373-380.
Christensen, M. N. 1965. Late Cenozoic deformation in the
central Coast Ranges of California. Geol. Soc. Am.
Bull. 76: 1105-1124.
Churkin, M., Jr. 1983. Tectonostratigraphic terranes of
Alaska and northeastern USSR—a record of collision
and accretion. In: M. Hashimoto and S. Uyeda (eds.),
Accretion Tectonics in the Circum-Pacific Regions.
Terra Scientific Publishing Company, Tokyo, pp. 37-
42.
Chyba, C. E, R J. Thomas, and K. J. Zahnle. 1993. The 1908
Tunguska explosion: atmospheric disruption of a
stony asteroid. Nature 361: 40-44.
Clarke, S. H., Jr., and G. A. Carver. 1992. Late Holocene tec-
tonics and paleoseismicity, southern Cascadia subduc-
tion zone. Science 255:188-192.
Clemens, S. C. and R. Tiedemann. 1997. Eccentricity forc-
ing of Pliocene-early Pliocene climate revealed in a
marine oxygen-isotope record. Nature 385: 801-804.
Clemens, W. A. and L. G. Nelms. 1993. Paleoecological im-
plications of Alaskan terrestrial vertebrate fauna in lat-
est Cretaceous time at high paleolatitudes. Geology 21:
503-506.
Clube, S. V. M. (ed.). 1989. Catastrophes and Evolution.
Cambridge University Press, Cambridge, U.K.
Cole, D. R. and H. C. Monger. 1994. Influence of atmo-
spheric CO2 on the decline of C4 plants during the last
deglaciation. Nature 368: 533-536.
Coleman, M. and K. Hodges. 1995. Evidence for Tibetan
Plateau uplift before 14 Myr ago from a new minimum
age for east-west extension. Nature 374: 49-52.
Colon, E 1959. The Life of Christopher Columbus by His
Son Ferdinand (translated and annotated by B. Keen).
Rutgers University Press, New Brunswick, NJ.
Cooper, A. and D. Penny. 1997. Mass survival of birds
across the Cretaceous-Tertiary boundary: molecular
evidence. Science 275: 1109-1113.
Covey, C. 1996. Sun-climate links. Science 272: 179.
and E. Barron. 1988. The role of ocean heat trans-
port in climatic change. Earth-Sci. Rev. 24: 429-445.
-, S. L. Thompson, P. R. Weissman, and M. C. Mac-
Cracken. 1994. Global climatic effects of atmospheric
dust from an asteroid or comet impact on Earth. Global
Planet. Change 9: 263-273.
Craig, R. G. 1987. Dynamics of a Missoula flood. In: L.
Meyer, and D. Nash (eds.), Eighteenth Annual Geomor-
phology Symposium: Catastrophic Flooding. Allen
and Unwin, London, pp. 305-332.
Croll, J. 1875. Climate and time in their geological rela-
tions—a theory of secular changes of the Earth's cli-
mate. D. Appleton, New York.
Cross, A. T. and R. E. Taggart. 1982. Causes of short-term
sequential changes in fossil plant assemblages: some
considerations based on a Miocene flora of the north-
west United States. Ann. Missouri Bot. Card. 69:
676-734.
Crowley, T. J. 1991. Utilization of paleoclimate results to

validate projections of a future greenhouse warming.
In: M. E. Schlesinger (ed.), Greenhouse-Gas-Induced
Climatic Change: A Critical Appraisal of Simulations
and Observations. Elsevier Science Publishers, Am-
sterdam, pp. 35-45.
. 1993. Geological assessment of the greenhouse ef-
fect. Bull. Am. Meterol. Soc. 74: 2363-2373.
, K.-Y. Kim, J. G. Mengel, and D. A. Short. 1992. Mod-
eling 100,000-year climate fluctuations in pre-
Pleistocene time series. Science 255: 705-707.
and G. R. North. 1988. Abrupt climate change and
extinction events in Earth history. Science 240: 996-
1002.
and . 1990. Modeling onset of glaciation.
Ann. Glaciol. 14: 39-42.
and . 1991. Paleoclimatology. Oxford Mono-
graphs in Geology and Geophysics. Oxford University
Press, Oxford, U.K. Monograph 16.
Crutzen, R J. and M. O. Andreae. 1990. Biomass burning in
the tropics: impact on atmospheric chemistry and bio-
geochemical cycles. Science 250:1669-1677.
Curry, W. B. and D. W. Oppo. 1997. Synchronous, high-
frequency oscillations in tropical sea surface tempera-
tures and North Atlantic deep water production dur-
ing the last glacial cycle. Paleoceanography 12: 1-14.
Dacey, J. W. H., B. G. Drake, and M. J. Klug. 1994. Stimula-
tion of methane emission by carbon dioxide enrich-
ment of marsh vegetation. Nature 370: 47-49.
Dalrymple, G. B., G. A. Izett, L. W. Snee, and J. D.
Obradovich. 1993. 40Ar/39Ar age spectra and total-
fusion ages of tektites from Cretaceous-Tertiary
boundary sedimentary rocks in the Beloc Formation,
Haiti. U.S. Geol. Survey Bull. 2065: 1-20.
Dansgaard, W. et al. (+ 5 authors). 1984. North Atlantic cli-
matic oscillations revealed by deep Greenland ice
cores. In: J. E. Hansen and T. Takahashi (eds.), Climate
Processes and Climate Sensitivity. American Geophys-
ical Union, Washington, B.C. Geophys. Monogr. 29,
pp. 288-298.
et al. (+ 10 authors). 1993. Evidence for general in-
stability of past climate from a 250-kyr ice-core record.
Nature 364: 218-220.
Dawes, R R. and J. W. Kerr (eds.). 1982. Nares Strait and the
Drift of Greenland: A Conflict in Plate Tectonics. Med-
delelser 0m Gr0nland, Geoscience 8, Copenhagen.
Decker, R. W. and B. B. Decker. 1991. Mountains of Fire: the
Nature of Volcanoes. Cambridge University Press,
Cambridge, U.K.
del Moral, R. and D. M. Wood. 1993. Early primary succes-
sion on a barren volcanic plain at Mount St. Helens,
Washington. Am. J. Bot. 80: 981-991.
de Vernal, A., C. Hillaire-Marcel, and G. Bilodeau. 1996.
Reduced meltwater outflow from the Laurentide ice
margin during the Younger Dryas. Nature 381: 774-
777.
Dewey, J. E, I. G. Gass, G. B. Curry, N. B. W. Harris, and
A. M. C. Sengor (eds.). 1991. Allochthonous Terranes.
Cambridge University Press, Cambridge, U.K.
Diaz, H. E and V. Markgraf (eds.). 1992. El Nino—Histori-
cal and Paleoclimatic Aspects of the Southern Oscilla-
tion. Cambridge University Press, Cambridge, U.K.
Dixon, R. K. et al. (+ 5 authors). 1994. Carbon pools and
flux of global forest ecosystems. Science 263: 185-
190.
Cause and Effect 75
Dowlatabadi, H. and M. G. Morgan. 1993. Integrated as-
sessment of climate change. Science 259: 1813, 1932.
Dunnette, D. A. and R. J. O. O'Brien (eds.). 1992. The Sci-
ence of Global Change, the Impact of Human Activities
on the Environment. American Chemical Society,
Washington, D.C.
Eaton, G. P. 1986. A tectonic redefinition of the southern
Rocky Mountains. Tectonophysics 132:163-193.
Edwards, R. L., H. Cheng, M. T. Murrell, and S. J. Gold-
stein. 1997. Protactinium-231 dating of carbonates by
thermal ionization mass spectrometry; implications
for Quarternary climate change. Science 276: 782-
786.
and C. D. Gallup. 1993. Technical comments: Dat-
ing of the Devils Hole calcite vein. Science 259:
1626.
Eldholm, O., A. M. Myhre, and J. Thiede. 1994. Cenozoic
tectomagmatic events in the North Atlantic: potential
paleoenvironmental implications. In: M. C. Boulter
and H. C. Fisher (eds.), Cenozoic Plants and Climates
of the Arctic. NATO ASI Series 1: Global Environ-
mental Change. Springer-Verlag, Berlin. Vol. 27, pp.
35-55.
Emiliani, C. 1955. Pleistocene temperatures. J. Geol. 63:
538-578.
Epstein, S., R. Buchsbaum, H. A. Lowenstam, and H. C.
Urey. 1951. Carbonate-water isotopic temperature
scale. Bull. Geol. Soc. Am. 62: 417-426.
Faure, H. 1990. Changes in the global continental reservoir
of carbon. Palaeogeogr., Palaeoecol., Palaeoclimatol.
82: 47-52.
Fiedel, S. J. 1995. Technical comment: The GISP ice core
record of volcanism since 7000 B.C. Science 267: 256.
Fiocco, G., D. Fu'a, and G. Visconti. 1997. The Mount
Pinatubo eruption, effects on the atmosphere arid cli-
mate. Springer-Verlag, Berlin.
Fischer, A. G. 1982. Long-term climatic oscillations
recorded in stratigraphy. In: Geophysics Study Com-
mittee (eds.), Climate in Earth History. National Acad-
emy Press, Washington, D.C. pp. 97-104.
Fisher, M. J, et al. (+ 6 authors). 1994. Carbon storage by in-
troduced deep-rooted grasses in the South American
savannas. Nature 371: 236-238.
Fitzgerald, P. G., E. Stump, and T. E Redfield. 1993. Late
Cenozoic uplift of Denali and its relation to relative
plate motion and fault morphology. Science 259: 497-
499.
Flam, E 1993. Is Earth's future climate written in the stars?
Science 262: 1372-1373.
Flora of North America Editorial Committee (eds.). 1993.
Flora of North America North of Mexico. Oxford Uni-
versity Press, Oxford, U.K.
Foley, J. A., J. E. Kutzbach, M. T. Coe, and S. Levis. 1994.
Feedbacks between climate and boreal forests during
the Holocene epoch. Nature 371: 52-54.
Freeman, K. H. and J. M. Hayes. 1992. Fractionation of car-
bon isotopes by phytoplankton and estimates of ancient
CO2 levels. Global Biogeochem. Cycles 6:185-198.
Friis-Christensen, E. and K. Lassen. 1991. Length of the
solar cycle: an indicator of solar activity associated
with climate. Science 254: 698-700.
Gallup, C. D., R. L. Edwards, and R. G. Johnson. 1994. The
timing of high sea levels over the past 200,000 years.
Science 263: 796-800.
76 Late Cretaceous and Cenozoic History of North American Vegetation
Gates, D. M. 1993. Climatic Change and its Biological Con-
sequences. Sinauer Associates, Sunderland, MA.
Gates, W. L. 1982. Paleoclimatic modeling—a review with
reference to problems and prospects for the pre-
Pleistocene. In: Geophysics Study Committee (eds.),
Climate in Earth History. National Academy Press,
Washington, D.C. pp. 26-42.
Geophysics Study Committee (eds.). 1982. Climate in Earth
History. National Academy Press, Washington, D.C.
. 1990. Sea-Level Change. National Academy Press,
Washington, D.C.
Gersonde, R. et al. (+ 12 authors). 1997. Geological record
and reconstruction of the Late Pliocene impact of the
Eltanin asteroid in the southern ocean. Nature 390:
357-363.
Giddings, L. V, K. Y. Kaneshiro, and W. W. Anderson (eds.).
1989. Genetics, Speciation, and the Founder Effect.
Oxford University Press, Oxford, U.K.
Glynn, R W. (ed.). 1990. Global Ecological Consequences of
the 1982-83 El Nino-Southern Oscillation. Elsevier
Oceanography Series. Elsevier Science Publishers,
Amsterdam. Vol. 52.
and W. H. de Weerdt. 1991. Elimination of two
reef-building hydrocorals following the 1982-83 El
Nino warming event. Science 253: 69-71.
Gordon, R. G. and S. Stein. 1992. Global tectonics and
space geodesy. Science 256: 333-342.
Grace, J. et al. (+ 11 authors). 1995. Carbon dioxide up-
take by an undisturbed tropical rain forest in south-
west Amazonia, 1992 to 1993. Science 270: 778-
780.
Greenland Ice-Core Project (GRIP) Members. 1993. Climate
instability during the last interglacial period recorded
in the GRIP ice core. Nature 364: 203-207.
Grieve, R. A. E 1996. Meteorite impact craters of North
America. In: R. L. Bishop (ed.), Observer's Handbook
1997. Royal Astronomical Soc. Can., Toronto, pp.
185-188.
Grimm, E. C., G. L. Jacobson, Jr., W. A. Watts, B. Hansen,
and K. A. Maasch. 1993. A 50,000-year record of cli-
mate oscillations from Florida and its temporal cor-
relation with the Heinrich events. Science 261:
198-200.
Grootes, P. M., M. Stuiver, J. White, S. Johnsen, and J.
Jouzel. 1993. Comparison of oxygen isotope records
from the GISP2 and GRIP Greenland ice cores. Nature
366:552-554.
Haigh, J. D. 1996. The impact of solar variability on cli-
mate. Science 272: 981-984.
Hammer, C. U. 1984. Traces of Icelandic eruptions in the
Greenland ice sheet. Jokull 34: 51-65.
, H. B. Clausen, and W. Dansgaard. 1980. Greenland
ice sheet evidence of post-glacial volcanism and its
climatic impact. Nature 288: 230-235.
Handler, R 1984. Possible association of stratospheric
aerosols and El Nino type events. Geophys. Res. Lett.
11: 1121-1124.
Hansen, J. et al. (+ 7 authors). 1991. Regional greenhouse
climate effects. In: M. E. Schlesinger (ed.), Green -
house-Gas-Induced Climatic Change: A Critical Ap-
praisal of Simulations and Observations. Elsevier Sci-
ence Publishers, Amsterdam, pp. 211-229.
, A. Lacis, R. Ruedy, M. Sato, and H. Wilson. 1993.
How sensitive is the world's climate? Natl. Geogr. Res.
Explor. 9: 143-158.
Hanski, I., R Turchin, E. Korpimakl, and H. Henttonen.
1993. Population oscillations of boreal rodents: regula-
tion by mustelid predators leads to chaos. Nature 364:
232-235.
Hanson, B. 1993. Editorial—evolution of atmospheres.
Science 259: 875.
Harrison, T. M., R Copeland, W. S. E Kidd, and A. Yin.
1992. Raising Tibet. Science 255: 1663-1670.
Hashimoto, M. and S. Uyeda (eds.). 1983. Accretion Tec-
tonics in the Circum-Pacific Regions. Terra Scientific
Publishing Company, Tokyo.
Hays, J. D., J. Imbrie, and N. J. Shackleton. 1976. Variations
in the Earth's orbit: pacemaker of the ice ages. Science
194: 1121-1132.
Hecht, A. D. (ed.). 1985a. Paleoclimate Analysis and Mod-
eling. Wiley-Interscience, New York.
. 1985b. Paleoclimatology: a retrospective of the past
20 years. In: A. D. Hecht (ed.), Paleoclimate Analysis
and Modeling. Wiley-Interscience, New York. pp.
1-25.
Heinrich, H. 1988. Origin and consequences of cyclic ice
rafting in the northeast Atlantic Ocean during the past
130,000 years. Quatern. Res. 29: 142-152.
Hide, R. and J. O. Dickey. 1991. Earth's variable rotation.
Science 253: 629-637.
Hildebrand, A. R. and W. V Boynton. 1990. Proximal Creta-
ceous-Tertiary boundary impact deposits in the Carib-
bean. Science 248: 843-847.
, M. Pilkington, M. Connors, C. Ortiz-Aleman, and
R. E. Chavez. 1995. Size and structure of the Chicxu-
lub crater revealed by horizontal gravity gradients and
cenotes. Nature 376: 415-417.
Hill, R. I., I. H. Campbell, G. E Davies, and R. W. Griffiths.
1992. Mantle plumes and continental tectonics. Sci-
ence 256: 186-193.
Hoffecker, J. E, W. R. Powers, and T. Goebel. 1993. The col-
onization of Beringia and the peopling of the New
World. Science 259: 46-53.
Hoffer, J. M., E Gomez, and R Muela. 1982. Eruption of El
Chichon volcano, Chiapas, Mexico, 28 March to 7
April 1982. Science 218: 1307-1308.
Hofmann, D. J. and J. M. Rosen. 1983. Sulfuric acid drop-
let formation and growth in the stratosphere after
the 1982 eruption of El Chichon. Science 222: 325-
327.
Holden, C. 1992. Energizing El Nino? Science 256: 741.
Hooper, R R. 1982. The Columbia River basalts. Science
215: 1463-1468.
Hopkins, D. M., J. V. Matthews, Jr., C. E. Schweger, and S. B.
Young (eds.). 1982. Paleoecology of Beringia. Aca-
demic Press, New York.
Houghton, J. T, B. A. Callander, and S. K. Varney. 1992. Cli-
mate Change 1992. IPCC (Intergovernmental Panel on
Climate Change) Supplement. Cambridge University
Press, Cambridge, U.K.
, G. J. Jenkins, and J. J. Ephraums (eds.). 1990. Cli-
mate Change: The IPCC Scientific Assessment. Cam-
bridge University Press, Cambridge, U.K.
Howell, D. G. 1989. Tectonics of Suspect Terranes, Moun-
tain Building and Continental Growth. Chapman and
Hall, New York.

Hoyt, D. V. and K. H. Schatten. 1997. The Role of the Sun in
Climate Change. Oxford University Press, Oxford, U.K.
Hsu, K. 1992. Challenger at Sea, A Ship that Revolution-
ized Earth Science. Princeton University Press, Prince-
ton, NJ.
Huber, N. K. 1981. Amount and Timing of Late Cenozoic
Uplift and Tilt of the Central Sierra Nevada, Califor-
nia—Evidence from the Upper San Joaquin River
Basin. U.S. Geological Survey, Denver, CO. Profes-
sional Paper 1197, pp. 1-28.
Huggett, R. J. 1990. Catastrophism: Systems of Earth His-
tory. Edward Arnold, London.
Hughen, K. A., J. T. Overpeck, L. C. Peterson, and S. Trum-
bore. 1996. Rapid climate changes in the tropical At-
lantic region during the last deglaciation. Nature 380:
51-54.
Hulten, E. 1937. Outline of the History of Arctic and Boreal
Biota during the Quaternary Period. Bokforlags Ak-
tiebolaget Thule, Stockholm.
Imbrie, J., A. C. Mix, and D. G. Martinson. 1993. Mi-
lankovitch theory viewed from Devils Hole. Nature
363: 531-533.
et al. (+ 8 authors). 1984. The orbital theory of Pleis-
tocene climate: support from a revised chronology of
the marine 18O record. In: A. L. Berger, J. Imbrie, J.
Hays, G. Kukla, and B. Saltzman (eds.), Milankovitch
and Climate: Understanding the Response to Astro-
nomical Forcing. Reidel, Dordrecht, pp. 269-305.
et al. (+17 authors). 1992. On the structure and ori-
gin of major glaciation cycles 1. Linear responses to
Milankovitch forcing. Paleoceanography 7: 701-738.
et al. (+ 18 authors). 1993. On the structure and ori-
gin of major glaciation cycles 2. The 100,000-year
cycle. Paleoceanography 8: 699-735.
Izett, G. A., W. A. Cobban, J. D. Obradovich, and M. J. Kunk.
1993. The Manson impact structure: 40Ar/39Ar age and
its distal impact ejecta in the Pierre Shale in southeast-
ern South Dakota. Science 262: 729-732.
, G. B. Dalrymple, and L. W. Snee. 1991. 40Ar/39Ar
age of Cretaceous-Tertiary boundary tektites from
Haiti. Science 252:1539-1542.
Janecek, T. R. 1985. Eolian sedimentation in the northwest-
ern Pacific Ocean: a preliminary examination of the
data from Deep Sea Drill. Proj. sites 576 and 578. Ini-
tial Rep. Deep Sea Drilling Project 86: 589-603.
Jansa, L. E 1993. Cometary impacts into ocean: their recog-
nition and the threshold constraint for biological ex-
tinctions. Palaeogeogr., Palaeoclimatol., Palaeoecol.
104: 271-286.
Jin, E-E, J. D. Neelin, and M. Ghil. 1994. El Nino on the
Devil's staircase: annual subharmonic steps to chaos.
Science 264: 70-72.
Johnsen, S. J. et al. (+ 9 authors). 1992. Irregular glacial in-
terstadials recorded in a new Greenland ice core. Na-
ture 359: 311-313.
Johnson, R. G., and S.-E. Lauritzen. 1995. Hudson Bay-
Hudson Strait jb'kulhlaups and Heinrich events: a hy-
pothesis. Palaeogeogr., Palaeoclimatol., Palaeoecol.
117: 123-137.
Jones, D. L., D. G. Howell, P. J. Coney, and J. W. H. Monger.
1983. Recognition, character, and analysis of tectonos-
tratigraphic terranes in western North America. In: M.
Hashimoto and S. Uyeda (eds.), Accretion Tectonics in
Cause and Effect 77
the Circum-Pacific Regions. Terra Scientific Publishing
Company, Tokyo, pp. 21-35.
-, N. J. Silberling, and J. Hillhouse. 1977. Wrangel-
lia—a displaced terrane in northwestern North Amer-
ica. Can. J. Earth Sci. 14: 2565-2577.
Jouzel, J. et al. (+ 16 authors). 1993. Extending the Vostok
ice-core record of palaeoclimate to the penultimate
glacial period. Nature 364: 407-412.
Kareiva, P M., J. G. Kingsolver, and R. B. Huey (eds.). 1993.
Biotic Interactions and Global Change. Sinauer Asso-
ciates, Sunderland, MA.
Kaufmann, R. K. and D. I. Stern. 1997. Evidence for human
influence on climate from hemispheric temperature
relations. Nature 388: 39-44.
Keeling, R. E, S. C. Piper, and M. Heimann. 1996. Global
and hemispheric CO2 sinks deduced from changes in
atmospheric O2 concentration. Nature 381: 218-221.
Keigwin, L. D., W. B. Curry, S. J. Lehman, and S. Johnsen.
1994. The role of the deep ocean in North Atlantic cli-
mate change between 70 and 130 Kyr ago. Nature 371:
323-326.
Keir, R. 1992. Packing away carbon isotopes. Nature 357:
445-446.
Kelly, R M. and C. B. Sear. 1984. Climatic impact of explo-
sive volcanic eruptions. Nature 311: 740-743.
and M. L. Wigley. 1992. Solar cycle length, green-
house forcing and global climate. Nature 360:
328-330.
Kennedy, J. A. and S. C. Brassell. 1992. Molecular records
of twentieth-century El Nino events in laminated sed-
iments from the Santa Barbara Basin. Nature 357:
62-64.
Kerr, R. A. 1995. A fickle sun could be altering Earth's cli-
mate after all. Science 269: 633.
. 1996a. A new dawn for sun-climate links? Science
271: 1360-1361.
-. 1996b. Ice rhythms: core reveals a plethora of cli-
mate cycles. Science 274: 499-500.
Kiehl, J. T. and B. P. Briegleb. 1993. The relative roles of sul-
fate aerosols and greenhouse gases in climate forcing.
Science 260: 311-314.
King, P. B. 1977. The Evolution of North America. Prince-
ton University Press, Princeton, NJ.
Kotra, J. P., D. L. Finnegan, W. H. Zoller, M. A. Hart, and J. L.
Moyers. 1983. El Chichon: composition of plume gases
and particles. Science 222:1018-1021.
Kring, D. A. and W. V. Boynton. 1992. Petrogenesis of an
augite-bearing melt rock in the Chicxulub structure
and its relationship to K/T impact spherules in Haiti.
Nature 358: 141-144. (See also discussion in Nature
362: 503-504 and 363: 503-504.)
Krogh, T. E., S. L. Kamo, V. L. Sharpton, L. E. Marin, and A.
R. Hildebrand. 1993. U-Pb ages of single shocked zir-
cons linking distal K/T ejecta to the Chicxulub crater.
Nature 366: 731-734.
Ku, T.-L., M. Ivanovich, and S. Luo. 1990. U-series dating of
last interglacial high sea stands: Barbados revisited.
Quatern. Res. 33: 129-147.
Kuchler, A. W. 1964. Manual to Accompany the Map Po-
tential Natural Vegetation of the Conterminous United
States. American Geographical Society, New York.
Special Publ. 36.
Kutzbach, J., G. Bonan, J. Foley, and S. P. Harrison. 1996.
78 Late Cretaceous and Cenozoic History of North American Vegetation
Vegetation and soil feedbacks on the response of the
African monsoon to orbital forcing in the Early to Mid-
dle Holocene. Nature 384: 623-626.
-, E J. Guetter, W. E Ruddiman, and W. L. Prell. 1989.
Sensitivity of climate to late Cenozoic uplift in south-
ern Asia and the American west: numerical experi-
ments. J. Geophys. Res. 94: 18393-18407.
LaMarche, V C., Jr., and K. K. Hirschboeck. 1984. Frost
rings in trees as records of major volcanic eruptions.
Nature 307: 121-126.
Lehman, S. J. and L. D. Keigwin. 1992. Sudden changes in
North Atlantic circulation during the last deglaciation.
Nature 356: 757-762.
Leinen, M. and G. R. Heath. 1981. Sedimentary indicators
of atmospheric activity in the northern hemisphere
during the Cenozoic. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 36: 1-21.
Levine, J. S. (ed.). 1991. Global Biomass Burning: Atmo-
spheric, Climatic, and Biospheric Implications. MIT
Press, Cambridge, MA.
Lewin, R. 1992. Complexity: Life at the Edge of Chaos.
MacMillan, New York.
Li, X., H. Maring, D. Savoie, K. Voss, and J. Prospero. 1996.
Dominance of mineral dust in aerosol light-scattering
in the North Atlantic trade winds. Nature 380: 416-
419.
List, R. J. 1951. Smithsonian Meterological Tables, Smith-
sonian Miscellaneous Collections. Smithsonian Insti-
tution Press, Washington, D.C. Vol. 114, 6th revised
edition.
Liu, H.-S. 1992. Frequency variations of the Earth's obliq-
uity and the 100-kyr ice-age cycles. Nature 358: 397-
399.
Ludwig, K. R., K. R. Simmons, I. J. Winograd, B. J. Szabo,
and A. C. Riggs. 1993. Technical comments: dating of
the Devils Hole calcite vein, response. Science 259:
1626-1627.
Lutjeharms, J. R. E. and H. R. Valentine. 1991. Sea-level
changes: consequences for the southern hemisphere.
Climatic Change 18: 317-337.
MacGiniiie, H. D. 1953. Fossil Plants of the Florissant Beds,
Colorado. Carnegie Institute of Washington, D.C. Con-
trib. Paleontol. Publ. 599.
Macilwain, C. 1995. El Nino centre is forecasting "first."
Nature 378: 228.
Maier-Reimer, E., U. Mikolajewicz, and T. J. Crowley. 1990.
Ocean general circulation model sensitivity experi-
ment with an open Central American isthmus. Paleo-
ceanography 5: 349-366.
Majumdar, S. K., L. S. Kalkstein, B. M. Yarnal, E. W. Miller,
and L. M. Rosenfeld (eds.). 1992. Global Climate
Change: Implications, Challenges and Mitigation Mea-
sures. Pennsylvania Academy of Science, Philadel-
phia, PA.
Manabe, S. and J. Broccoli. 1990. Mountains and arid cli-
mates of middle latitudes. Science 247: 192-195.
and R. J. Stouffer. 1995. Simulation of abrupt cli-
mate change induced by freshwater input to the North
Atlantic Ocean. Nature 378: 165-167.
and T. B. Terpstra. 1974. The effects of mountains on
the general circulation of the atmosphere as identified
by numerical experiments. J. Atmos. Sci. 31: 3-42.
Marine Geology. 1996. 131 (April): whole issue.
Marshall, C. R. and P. D. Ward. 1996. Sudden and gradual
molluscan extinctions in the latest Cretaceous of west-
ern European Tethys. Science 274: 1360-1363.
Mass, C. E and D. A. Portman. 1989. Major volcanic erup-
tions and climate: a critical evaluation. J. Glim. 2:
566-593.
Matthews, J. V, Jr., and L. E. Ovenden. 1990. Late Tertiary
plant macrofossils from localities in Arctic/subarctic
North America: a review of the data. Arctic 43:
364-392.
Maurrasse, E and G. Sen. 1991. Impacts, tsunamis, and the
Haitian Cretaceous-Tertiary boundary layer. Science
252: 1690-1693.
Mayewski, P A. et al. (+ 13 authors). 1996. Climate change
during the last deglaciation in Antarctica. Science 272:
1636-1638.
McCormick, M. R, L. W. Thomason, and C. R. Trepte. 1995.
Atmospheric effects of the Mt Pinatuba eruption. Na-
ture 373: 399-404.
McGuffie, K., A. Henderson-Sellers, H. Zhang, T. B. Dur-
bidge, and A. J. Pitman. 1995. Global climate sensitiv-
ity to tropical deforestation. Global Planet. Change 10:
97-128.
Mclntyre, A., and B. Molfino. 1996. Forcing of Atlantic
equatorial and subpolar millennial cycles by preces-
sion. Science 274: 1867-1870.
McKenna, M. C. 1972. Was Europe connected directly to
North America prior to the middle Eocene? In:
T. Dobzhansky, M. K. Hecht, and W. C. Steere (eds.),
Evolutionary Biology. Appleton-Century-Crofts, New
York. Vol. 6., pp. 179-189.
. 1983. Holarctic landmass rearrangement, cosmic
events, and Cenozoic terrestrial organisms. Ann. Mis-
souri Bot. Garden 70: 459-489.
McManus, J. E et al. (+ 5 authors). 1994. High-resolution
climate records from the North Atlantic during the last
interglacial. Nature 371: 326-329.
Meese, D. A. et al. (+ 8 authors). 1994. The accumulation
record from the GISP2 core as an indicator of climate
change throughout the Holocene. Science 266: 1680-
1682.
Mesolella, K. J., R. K. Matthews, W. S. Broecker, and D. L.
Thurber. 1969. The astronomical theory of climatic
change: Barbados data. J. Geol. 77: 250-274.
Meyer, H. W. 1992. Lapse rates and other variables applied
to estimating paleoaltitudes from fossil floras. Palaeo-
geogr., Palaeoclimatol., Palaeoecol. 99: 71-99.
Milankovitch, M. 1920. Theorie mathematique des phe-
nomenes thermiques produits par la radiation solaire.
Gauthier-Villars, Academie Yugoslave des Sciences et
des Arts de Zagreb, Yugoslavia.
. 1930. Mathematische klimalehre und astronomis-
che theorie de klimschwankungen. In: W. Koppen and
R. Geiger (eds.), Handbuch der Klimatologie, 1. Teil A.
Borntraeger, Berlin.
-. 1941. Canon of Insolation and the Ice Age Problem.
Koninglich Serbische Akademie, Beogra, Serbia. [Eng-
lish translation of the Israel Program for Scientific
Translation, published for the U.S. Dept. Commerce
and the National Science Foundation].
Miller, K. G., R. G. Fairbanks, and G. S. Mountain. 1987.
Tertiary oxygen isotope synthesis, sea level history,
and continental margin erosion. Paleoceanography 2:
1-19.
Minnis, R et al. (+ 6 authors). 1993. Radiative climate fore-

ing by the Mount Pinatubo eruption. Science 259:
1411-1415.
Mintz, Y. and G. Dean. 1952. The Observed Mean Field of
Motion of the Atmosphere. Air Force Cambridge Re-
search Center, Cambridge, U.K. Geophys. Res. Paper 17.
Molnar, P. and P. England. 1990. Late Cenozoic uplift of
mountain ranges and global climate change: chicken
or egg? Nature 346: 29-34.
Morell, V. 1993. How lethal was the K-T impact? Science
261: 1518-1519.
Morgan, M. E., J. D. Kingston, and B. D. Marino. 1994. Car-
bon isotopic evidence for the emergence of C4 plants
in the Neogene from Pakistan and Kenya. Nature 367:
162-165.
Mull, C. G. 1979. Nanushuk Group deposition and late
Mesozoic structural evolution of the central and west-
ern Brooks Range and Arctic Slope. In: T. S. Ahlbrandt
(ed.), Preliminary Geologic, Petrologic, and Paleonto-
logic Results of the Study of Nanushuk Group Rocks,
North Slope, Alaska. U.S. Geological Survey, Denver,
CO. Circular 794, pp. 5-13.
. 1985. Cretaceous tectonics, depositional cycles,
and the Nanushuk Group, Brooks Range and Arctic
Slope, Alaska. U.S. Geol. Surv. Bull. 1614: 7-36.
Nature. 1992. 357 (May 28): whole issue.
Nature. 1993. 361:123-124.
Nature. 1994. 371: 112-113.
Neftel, A., H. Oeschger, J. Schwander, B. Stauffer, and R.
Zumbrunn. 1982. Ice core sample measurements give
atmospheric CO2 content during the past 40,000 yr.
Nature 295: 220-223.
NRG Panel on Climatic Variation, U.S. Committee for
GARP. 1975. Understanding Climatic Change—A Pro-
gram for Action. National Academy Press, Washing-
ton, D.C.
Oechel, W. C. et al. (+ 5 authors). 1993. Recent change of
Arctic tundra ecosystems from a net carbon dioxide
sink to a source. Nature 361: 520-523.
Oldow, J. S., A. W. Bally, H. G. Ave Lallemant, and W. R
Leeman. 1989. Phanerozoic evolution of the North
American Cordillera; United States and Canada. In:
A. W. Bally and A. R. Palmer (eds.), The Geology of
North America; An Overview. DNAG (Decade of North
American Geology), The Geology of North America.
Geological Society of America, Boulder, CO. Vol. A,
pp. 139-232.
Olsen, P. E. 1986. A 40-million-year lake record of early
Mesozoic orbital climatic forcing. Science 234:
842-848.
and D. V. Kent. 1996. Milankovitch climate forcing
in the tropics of Pangaea during the Late Triassic.
Palaeogeogr., Palaeoclimatol., Palaeoecol. 122:1-26.
Oppenheimer, D. et al. (+ 19 authors). 1993. The Cape Men-
docino, California earthquakes of April 1992: subduc-
tion at the triple junction. Science 261: 433-438.
Oppo, D. W. and S. J. Lehman. 1993. Mid-depth circulation
of the subpolar North Atlantic during the last glacial
maximum. Science 259:1148-1152.
Oberbeck, V. R. and R. L. Mancinelli. 1994. Asteroid im-
pacts, microbes, and the cooling of the atmosphere.
BioScience 44:173-177.
Overpeck, J. T. 1992. Review: Global biomass burning: at-
mospheric, climatic, and biospheric implications, J. S.
Levine (ed.). Nature 356: 670.
Cause and Effect 79
, E J. Bartlein, and T. Webb III. 1991. Potential magni-
tude of future vegetation change in eastern North Amer-
ica: comparisons with the past. Science 254: 692-695.
Paillard, D. and L. Labeyrie. 1994. Role of the thermohaline
circulation in the abrupt warming after Heinrich
events. Nature 372:162-164.
Park, J., S. L. D'Hondt, J. W. King, and C. Gibson. 1993. Late
Cretaceous precessional cycles in double time: a
warm-Earth Milankovitch response. Science 261:1431-
1434.
Parker, D. E. and J. L. Brownscombe. 1983. Stratospheric
warming following the El Chichon volcanic eruption.
Nature 301: 406-408.
Partridge, T. C., G. C. Bond, C. J. H. Hartnady, P. B. deMeno-
cal, and W. E Ruddiman. 1995. Climatic effects of late
Neogene tectonism and volcanism. In: E. S. Vrba,
G. H. Denton, T. C. Partridge, and L. H. Burckle (eds.),
Paleoclimate and Evolution, with Emphasis on Human
Origins. Yale University Press, New Haven, CT. pp.
8-23.
Paterson, W. S. B. and C. U. Hammer. 1987. Ice core and other
glaciological data. In: W. E Ruddiman and H. E. Wright,
Jr. (eds.), North America and Adjacent Oceans during
the Last Deglaciation. The Geology of North America,
DNAG (Decade of North American Geology). Geological
Society of America, Boulder, CO. Vol. K-3, pp. 91-109.
Pedolosky, J. 1990. The dynamics of the oceanic subtropi-
cal gyres. Science 248: 316-322.
Peters, R. L. and T. E. Lovejoy (eds.). 1992. Global Warming
and Biological Diversity. Yale University Press, New
Haven, CT.
Philander, S. G. H. 1983. El Nino southern oscillation phe-
nomena. Nature 302: 295-301.
Poag, C. W. 1997. The Chesapeake Bay bolide impact: a
convulsive event in Atlantic coastal plain evolution.
Sedimentary Geol. 108: 45-90.
Polley, H. W, H. B. Johnson, B. D. Marino, and H. S.
Mayeux. 1993. Increase in C3 plant water-use effi-
ciency and biomass over glacial to present CO2 con-
centrations. Nature 361: 61-64.
Prell, W. L. and J. E. Kutzbach. 1992. Sensitivity of the In-
dian monsoon to forcing parameters and implications
for its evolution. Nature 360: 647-652.
Price, J. E et al. (+ 8 authors). 1993. Mediterranean outflow
mixing and dynamics. Science 259: 1277-1282.
Quade, J., T. E. Cerling, and J. R. Bowman. 1989. Develop-
ment of Asian monsoon revealed by marked ecological
shift during the latest Miocene in northern Pakistan.
Nature 342: 163-166.
Quinn, W. H., V. T. Neal, and S. E. Antunez de Mayolo.
1987. El Nino occurrences over the past four and a half
centuries. J. Geophys. Res. 92:14449-14461.
, D. O. Zopf, K. S. Short, and R. T. W. Kuo Yang. 1978.
Historical trends and statistics of the southern oscilla-
tion, El Nino, and Indonesian droughts. Fish. Bull. 76:
663-678.
Rahmstorf, S. 1994. Rapid climate transitions in a coupled
ocean-atmosphere model. Nature 372: 82-85.
. 1995. Bifurcations of the Atlantic thermohaline cir-
culation in response to changes in the hydrological
cycle. Nature 378: 145-149.
Ramanathan, V. et al. (+ 7 authors). 1995. Warm pool heat
budget and shortwave cloud forcing: a missing phys-
ics? Science 267: 499-503.
80 Late Cretaceous and Cenozoic History of North American Vegetation
Rampino, M. R., S. Self, and R. B. Stothers. 1988. Volcanic
winters. Annu. Rev. Earth Planet. Sci. 16: 73-99.
Rathburn, S. L. 1993. Pleistocene cataclysmic flooding
along the Big Lost River, east central Idaho. Geomor-
phology 8: 305-319.
Raup, D. M. 1981. Introduction: what is crisis? In: M. H.
Nitecki (ed.), Biotic Crises in Ecological and Evolu-
tionary Time. Academic Press, New York. pp. 1-12.
Raymo, M. E. and W. E Ruddiman. 1992. Tectonic forcing
of late Cenozoic climate. Nature 359: 117-122.
. 1993. Scientific correspondence: Cooling in the
late Cenozoic. Nature 361:124.
Raynaud, D. et al. (+ 5 authors). 1993. The ice record of
greenhouse gases. Science 259: 926-934.
Rea, D. K., M. Leinen, and T. R. Janecek. 1985. Geologic ap-
proach to the long-term history of atmospheric circula-
tion. Science 227: 721-725.
Repenning, C. A. and E. M. Brouwers. 1992. Late Pliocene-
early Pleistocene ecologic changes in the Arctic Ocean
borderland. U.S. Geol. Surv. Bull. 2036: 1-37.
Retallack, G. J. 1996. Acid trauma at the Cretaceous -
Tertiary boundary in eastern Montana. GSA Today 6:
1-7.
Rind, D., C. Rosenzweig, and R. Goldberg. 1992. Modelling
the hydrological cycle in assessments of climatic
change. Nature 358: 119-122.
Robin, E., L. Froget, C. Jehanno, and R. Rocchia. 1993. Evi-
dence for a K/T impact event in the Pacific Ocean. Na-
ture 363: 615-617.
Robock, A. and J. Mao. 1992. Winter warming from large
volcanic eruptions. Geophys. Res. Lett. 19: 2405-
2408.
Rozanski, K., L. Araguas-Araguas, and R. Gonfiantini.
1992. Relation between long-term trends of oxygen-18
isotope composition of precipitation and climate. Sci-
ence 258: 981-985.
Rubin, E. S. et al. (+ 6 authors). 1992. Realistic mitigation
options for global warming. Science 257: 148-149;
261-266.
Ruddiman, W. E 1997. Tectonic Uplift and Climatic
Change. Plenum, New York.
Ruddiman, W. E and J. E. Kutzbach. 1989. Forcing of late
Cenozoic northern hemisphere climate by plateau up-
lift in southern Asia and the American west. J. Geo-
phys. Res. 94: 18409-18427.
. 1990. Late Cenozoic plateau uplift and climate
change. Trans. R. Soc. Edinburgh: Earth Sci. 81: 301-
314.
Ruddiman, W. E and A. Mclntyre. 1976. Northeast Atlantic
paleoclimatic changes over the past 600,000 years. In:
R. M. Cline and J. D. Hays (eds.), Investigation of Late
Quaternary Paleoceanography and Paleoclimatology.
Geological Society of America, Boulder, CO. Memoir
145, pp. 111-146.
. 1984. Ice-age thermal response and climatic role of
the surface Atlantic Ocean, 40° to 63° N. Geol. Soc.
Am. Bull. 95: 381-396.
Ruddiman, W. E, W. L. Prell, and M. E. Raymo. 1989. Late
Cenozoic uplift in southern Asia and the American
west: rationale for general circulation modeling exper-
iments. J. Geophys. Res. 94: 18379-18391.
Ruelle, D. 1991. Chance and Chaos. Princeton University
Press, Princeton, NJ.
Sacks, P. E. and D. T. Secor, Jr. 1990. Kinematics of late Pa-
leozoic continental collision between Laurentia and
Gondwana. Science 250: 1702-1705.
Sagan, C. and S. J. Ostro. 1994. Dangers of asteroid deflec-
tion. Nature 368: 501.
Salati, E., T. E. Lovejoy, and R B. Vose. 1983. Precipitation
and water recycling in tropical rain forests with spe-
cial reference to the Amazon Basin. Environmentalist
3: 67-72.
Salati, E. and P. B. Vose. 1984. Amazon Basin: a system in
equilibrium. Science 225: 129-138.
Saltzman, B. and M. Verbitsky. 1994. CO2 and glacial cy-
cles. Nature 367: 419.
Sanford, R. L., Jr., J. Saldarriaga, K. E. Clark, C. Uhl, and R.
Herrera. 1985. Amazon rain-forest fires. Science 277:
53-55.
Savage, J. C. and M. Lisowski. 1991. Strain measurements
and the potential for a great subduction earthquake off
the coast of Washington. Science 252:101-103.
Savin, S. M. and R. G. Douglas. 1985. Sea level, climate,
and the Central American land bridge. In: E G. Stehli
and S. D. Webb (eds.), The Great American Biotic In-
terchange. Plenum, New York. pp. 303-324.
Schlesinger, M. E. and N. Ramankutty. 1992. Implications
for global warming of intercycle solar irradiance vari-
ations. Nature 360: 330-333.
Schmidlin, T. W. 1989. Climatic summary of snowfall and
snow depth in the Ohio snowbelt at Chardon. Ohio J.
Sci. 89: 101-108.
Schneider, R. and M. Little. 1997. Southern and northern
hemisphere climatic changes: synchronous or not?
PAGES 5:12.
Schneider, S. H. 1994. Detecting climatic change signals:
are there any "fingerprints"? Science 263: 341-347.
Schwarzacher, W. 1993. Cyclostratigraphy and the Mi-
lankovitch Theory. Elsevier Science Publishers, Am-
sterdam.
Science. 1988. 241: 717-724.
Science. 1992. 258: 1611-1630.
Science. 1993. 259 (February 12): cover.
Scientific Committee on Oceanic Research (SCOR), Work-
ing Group 55. 1983. Prediction of "El Nino." SCOR
Proc. (Paris) 19: 47-51.
Scott, G. R. 1975. Cenozoic surfaces and deposits in the
southern Rocky Mountains. In: B. E Curtis (ed.), Ceno-
zoic History of the Southern Rocky Mountains. Geo-
logical Society of America, Boulder, CO. Memoir 144,
pp. 227-248.
Scott, J. V, D. L. Rabinowitz, and B. G. Marsden. 1991. Near
miss of the Earth by a small asteroid. Nature 354:
287-289.
Severinghaus, J. P., T. Sowers, E. J. Brook, R. B. Alley, and
M. L. Bender. 1998. Timing of abrupt climate change at
the end of the Younger Dryas interval from thermally
fractionated gases in polar ice. Nature 391: 141-146.
Shackleton, N. J., A. Berger, and W. R. Peltier. 1990. An al-
ternative astronomical calibration of the lower Pleis-
tocene timescale based on OOP site 677. Trans. R. Soc.
Edinburgh: Earth Sci. 81: 251-261.
Shackleton, N. J., M. A. Hall, J. Line, and C. Shuxi. 1983.
Carbon isotope data in core V19-30 confirm reduced
carbon dioxide concentration in the ice age atmo-
sphere. Nature 306: 319-322.
Shackleton, N. J. and J. R Kennett. 1975. Late Cenozoic oxy-
gen and carbon isotopic changes at DSDP site 284: im-

plications for glacial history of the northern hemisphere.
Initial Rep. Deep Sea Drilling Project 29: 801-807.
Sharpton, V. L. and P. D. Ward (eds.). 1990. Global Catastro-
phies in Earth History. An Interdisciplinary Confer-
ence on Impacts, Volcanism, and Mass Mortality,
Snowbird, Utah, 1988. Geological Society of America,
Boulder, CO. Special Paper 247.
Sharpton, V. L. et al. (+ 5 authors). 1992. New links be-
tween the Chicxulub impact structure and the Creta-
ceous/Tertiary boundary. Nature 359: 819-821.
Sharpton, V. L. et al. (+ 9 authors). 1993. Chicxulub mul-
tiring impact basin: size and other characteristics de-
rived from gravity analysis. Science 261: 1564-
1567.
Sheehan, P. M., D. E. Fastovsky, R. G. Hoffmann, C. B.
Berghaus, and D. L. Gabriel. 1991. Sudden extinctions
of the dinosaurs: latest Cretaceous, upper Great Plains,
U.S.A. Science 254: 835-839.
Shukla, J. and Y. Mintz. 1982. Influence of land-surface
evapotranspiration on the Earth's climate. Science 215:
1498-1501.
Shukla, J., C. Nobre, and P. Sellers. 1990. Amazon defor-
estation and climatic change. Science 247: 1322-
1325.
Siegenthaler, U. and J. L. Sarmiento. 1993. Atmospheric
carbon dioxide and the ocean. Nature 365:119-125.
Sloan, L. C. and E. J. Barren. 1990. "Equable" climates dur-
ing Earth history? Geology 18: 489-492.
Slowey, N. C., G. M. Henderson, and W. B. Curry. 1996. Di-
rect U-Th dating of marine sediments from the two
most recent interglacial periods. Nature 383: 242-244.
Small, E. E. and R. S. Anderson. 1995. Geomorphically
driven late Cenozoic rock uplift in the Sierra Nevada,
California. Science 270: 277-280.
Smith, T. M. and H. H. Shugart. 1993. The transient re-
sponse of terrestrial carbon storage to a perturbed cli-
mate. Nature 361: 523-526.
Southon, J. R. and T. A. Brown. 1995. Technical Comment:
The GISP ice core record of volcanism since 7000 B.C.
Science 267: 256-257.
Stewart, I. 1989. Does God Play Dice?: The Mathematics of
Chaos. Blackwell, New York.
Stone, D. B., B. C. Panuska, and D. R. Packer. 1982. Paleolat-
itudes versus time for southern Alaska. J. Geophys.
Res. 87: 3697-3707.
Summerfield, M. A. and M. P. Kirkbride. 1992. Climate and
landscape response. Nature 355: 306. (See also Molnar
and England reply.)
Svetsov, V. V. 1996. Total ablation of the debris from the
1908 Tunguska explosion. Nature 383: 697-699.
Swisher C. C., Ill (+ 11 authors). 1992. Coeval 40Ar/39Ar
ages of 65.0 million years ago from Chicxulub Crater
melt rock and Cretaceous-Tertiary boundary tektites.
Science 257: 954-958.
Taggart, R. E. and A. T. Cross. 1980. Vegetation change in
the Miocene Succor Creek flora of Oregon and Idaho: a
case study in paleosuccession. In: D. L. Dilcher and
T. N. Taylor (eds.), Biostratigraphy of Fossil Plants.
Dowden, Hutchinson, & Ross, Stroudsburg, PA. pp.
185-210.
Taylor, K. C. et al. (+ 7 authors). 1993a. The "flickering
switch" of late Pleistocene climate change. Nature 361:
432-436.
Taylor, K. C. et al. (+ 9 authors). 1993b. Electrical conduc-
Cause and Effect 81
tivity measurements from the GISP 2 and GRIP Green-
land ice cores. Nature 366: 549-552.
Taylor, K. C. et al. (+ 12 authors). 1997. The Holocene-
Younger Dryas transition recorded at Summit, Green-
land. Science 278: 825-827.
Tegen, L, A. A. Lacis, and I. Fung. 1996. The influence on
climate forcing of mineral aerosols from disturbed
soils. Nature 380: 419-422.
Thelin, G. P. and R. J. Pike. 1991. Landforms of the conter-
minous United States—a digital shaded-relief por-
trayal. [Text] U.S. Geological Survey, Denver, CO. Mis-
eel. Invest. Ser. Map 1-2206.
Tilling, R. I. et al. (+ 5 authors). 1984. Holocene eruptive
activity of El Chichon volcano, Chiapas, Mexico. Sci-
ence 224: 747-749.
Trewartha, G. T. and L. H. Horn. 1980. An Introduction to
Climate. McGraw-Hill, New York. 5th edition.
Turner, S., C. Hawkesworth, J. Liu, N. Rogers, S. Kelley, and
P. van Calsteren. 1993. Timing of Tibetan uplift con-
strained by analysis of volcanic rocks. Nature 364:
50-54.
Tziperman, E., L. Stone, M. A. Cane, and H. Jarosh. 1994.
El Nino chaos: overlapping of resonances between the
seasonal cycle and the Pacific Ocean-atmosphere os-
cillator. Science 264: 72-74.
Urey, H. C. 1947. The thermodynamic properties of iso-
topic substances. J. Chem. Soc. 562-581.
, H. A. Lowenstam, S. Epstein, and C. R. McKinney.
1951. Measurement of paleotemperatures and temper-
atures of the Upper Cretaceous of England, Denmark,
and the southeastern United States. Bull. Geol. Soc.
Am. 62: 399-416.
Vail, R R. and J. Hardenbol. 1979. Sea-level changes during
the Tertiary. Oceanus 22: 71-79.
Van Houten, E B. 1964. Cyclic lacustrine sedimentation,
Upper Triassic Lockatong Formation, central New Jer-
sey and adjacent Pennsylvania. Bull. Kansas State
Geol. Survey 169: 497-531.
Volk, T. 1993. Scientific correspondence: Cooling in the
late Cenozoic. Nature 361:123.
Waitt, R. B., Jr., 1985. Case for periodic, colossal jokulh-
laups from Pleistocene Glacial Lake Missoula. Bull.
Geol. Soc. Am. 96:1271-1286.
Waldrop, M. M. 1992. Complexity: The Emerging Science
at the Edge of Order and Chaos. Simon & Schuster/
Viking, New York.
Wallace, R. E. (ed.). 1990. The San Andreas Fault System,
California. U.S. Geological Survey, Denver, CO.
Ward, P. D., J. M. Hurtado, J. L. Kirschvink, and K. L. Vero-
sub. 1997. Measurements of the Cretaceous paleolati-
tude of Vancouver Island: Consistent with the
Baja-British Columbia Hypothesis. Science 277:
1642-1645.
Webb, R. S., D. H. Rind, S. J. Lehman, R. J. Healy, and D. Sig-
man. 1997. Influence of ocean heat transport on the
climate of the last glacial maximum. Nature 385:
695-699.
West, E H. (ed.). 1996. American Beginnings: the prehistory
and palaeocology of Beringia. University of Chicago
Press, Chicago.
White, J. W. C., P. Ciais, R. A. Figge, R. Kenny, and V. Mark-
graf. 1994. A high-resolution record of atmospheric
CO2 content from carbon isotopes in peat. Nature 367:
153-156.
82 Late Cretaceous and Cenozoic History of North American Vegetation
Wigley, T. M. L. and P D. Jones. 1981. Detecting CO2-
induced climatic change. Nature 292: 205-208.
Williams, M. A. J., D. L. Dunkerley, R De Deckker, A. P. Ker-
shaw, and T. Stokes. 1993. Quaternary Environments.
Edward Arnold, London.
Winograd, I. J., B. J. Szabo, T. B. Coplen, and A. C. Riggs.
1988. A 250,000-year climatic record from Great Basin
vein calcite: implications for Milankovitch theory. Sci-
ence 242: 1275-1280.
Winograd, I. J. et al. (+ 7 authors). 1992. Continuous
500,000-year climate record from vein calcite in Devils
Hole, Nevada. Science 258: 255-260.
Winterer, E. L., T. M. Atwater, and R. W. Decker. 1989. The
northeast Pacific Ocean and Hawaii. In: A. W. Bally
and A. R. Palmer (eds.), The Geology of North Amer-
ica; an Overview. DNAG (Decade of North American
Geology). Geological Society of America, Boulder, CO.
Vol. A, pp. 265-298.
Wolfe, J. A. 1987. Late Cretaceous-Cenozoic history of de-
ciduousness and the terminal Cretaceous event. Paleo-
biology 13: 215-226.
Wood, C. A. and J. Kienle (eds.). 1990. Volcanoes of North
America: United States and Canada. Cambridge Uni-
versity Press, Cambridge, U.K.
Woodward, E 1.1987. Climate and Plant Distribution. Cam-
bridge University Press, Cambridge, U.K.
Woodwell, G. M. (+ 6 authors). 1983. Global deforestation:
contribution to atmospheric carbon dioxide. Science
222: 1081-1086.
World Meterological Organization [E Steinhauser (Tech.
Super.)!. 1979. Climatic Atlas of North and Central
America—Maps of Mean Temperature and Precipita-
tion. World Meteorological Organization, Geneva.
Zhao, M. and J. L. Bada. 1989. Extraterrestrial amino acids
in Cretaceous/Tertiary boundary sediments at Stevns
Klint, Denmark. Nature 339: 463-465.
Ziegler, R A. 1988. Evolution of the Arctic-North Atlantic
and the Western Tethys. American Association of Pe-
troleum Geologists, Tulsa, OK. Mem. 43.
Zielinski, G. A. et al. (+ 8 authors). 1994. Record of volcan-
ism since 7000 B.C. from the GISP2 Greenland ice core
and implications for the volcano-climate system. Sci-
ence 264: 948-952. (See responses by Fiedel, 1995;
Southon and Brown, 1995.)
General Readings
Adams, J. 1992. Paleoseismology: a search for ancient
earthquakes in Puget Sound. Science 258:1592-1593.
Albarede, E 1997. Isotopic tracers of past ocean circulation;
turning lead into gold. Science 277: 908-909.
Alley, R. B., and M. L. Bender. 1998. Greenland ice cores:
frozen in time. Sci. Am. 278: 80-85.
Alvarez, W. 1997. T. rex and the Crater of Doom. Princeton
University Press, Princeton, NJ.
and E Asaro. 1990. An extraterrestrial impact. Sci.
Am. 263: 78-84.
Amato, I. 1993. Coming in loud and clear—via chaos. Sci-
ence 261: 429.
Andreae, M. O. 1996. Raising dust in the greenhouse. Na-
ture 380: 389-390.
Anonymous. 1994. El Nino and climate prediction. Report
to the nation on our changing planet. Univ. Corp.
Atmos. Res. Spring.
Appenzeller, T. 1993a. Searching for clues to ancient car-
bon dioxide. Science 259: 908-909.
. 1993b. Ancient climate coolings are on thin ice.
Science 262: 1818-1819.
Ball, R 1992. Raising the roof of the world. Nature 357: 642.
Barnes-Svarney, R 1996. Asteroid, Earth Destroyer or New
Frontier? Plenum, New York.
Bender, M. 1996. A quickening on the uptake? Nature 381:
195-196.
Berner, R. A. and A. C. Lasaga. 1989. Modeling the geo-
chemical carbon cycle. Sci. Am. 260: 74-81.
Bigg, G. R. 1996. The Oceans and Climate. Cambridge Uni-
versity Press, Cambridge, U.K.
Blum, J. D. 1993. Zircon can take the heat. Nature 366: 718.
Bond, G. C. 1995. Climate and the conveyor. Nature 377:
383-384.
Boulton, G. S. 1993. Two cores are better than one. Nature
366: 507-508.
Briggs, J. C. 1991. A Cretaceous-Tertiary mass extinction?
BioScience 41: 619-624.
Broecker, W. S. 1987. The biggest chill. Natural History 96:
74-82.
. 1992. Upset for Milankovitch theory. Nature 359:
779-780.
. 1994. An unstable superconveyor. Nature 367:
414-415.
. 1995. The Glacial World According to Wally. Lam-
ont-Doherty Earth Observatory, Palisades, NY.
and G. H. Denton. 1990. What drives glacial cycles?
Sci. Am. 262: 49-56.
Burke, I. C. (+ 5 authors). 1991. Regional analysis of the
central Great Plains—sensitivity to climate variability.
BioScience 41: 685-692.
Carroll, M. R. 1997. Volcanic sulphur in the balance. Na-
ture 389: 543-544.
Charles, C. 1997. Cool tropical punch of the Ice Ages. Na-
ture 385: 681-683.
Clifton, H. E. and R. H. Dott, Jr. 1994. Sedimentology of the
K-T boundary. Science 264: 1829. (See also Keller,
1994; Kerr, 1994c.)
Committee on International Science's Task Force on Global
Biodiversity. 1989. Loss of Biological Diversity: A
Global Crisis Requiring International Solutions. A re-
port to the National Science Foundation. National
Science Foundation, Washington, D.C.
Cook, E A., L. D. Brown, and J. E. Oliver. 1980. The south-
ern Appalachians and the growth of continents. Sci.
Am. 243: 156-168.
Courtillot, V. E. 1990. A volcanic eruption. Sci. Am. 263:
85-92.
Crowley, T. J. 1996. Remembrance of things past: green-
house lessons from the geologic record. Consequences
2:3-12.
Culotta, E. 1993. Is the geological past a key to the (near)
future? Science 259: 906-908.
Dauber, R M. and R. A. Muller. 1996. The Three Big Bangs:
Comet Crashes, Exploding Stars, and the Creation of
the Universe. Addison-Wesley, Reading, MA.
Davies, G. E 1992. Plates and plumes: dynamos of the
Earth's mantle. Science 257: 493-494.
Decker, R. W. and B. B. Decker. 1981. The eruptions of
Mount St. Helens. Sci. Am. 244: 68-80.
. 1991. Mountains of Fire: The Nature of Volcanoes.
Cambridge University Press, Cambridge, U.K.

Deming, D. 1995. Climatic warming in North America:
analysis of borehole temperatures. Science 268:
1576-1577.
Gibbons, A. 1997. Did birds sail through the K-T extinc-
tion with flying colors? Science 275:1068.
Glantz, M. H. 1996. Currents of change, El Nino's impact
on climate and society. Cambridge University Press,
Cambridge, U.K.
Gordon, A. L. 1996. Communication between oceans. Na-
ture 382: 399-400.
Gould, S. J. 1967. Is uniformitarianism useful? Geol. Ed. 15:
149-150.
. 1978. The great scablands debate. Nat. Hist. 87:
12-18.
Holden, C. 1992a. Pack rats' liquid legacy. Science 255:
155.
. 1992b. Cracking secrets in Greenland ice. Science
257:161.
. 1996. Environment award honors the big chill. Sci-
ence 272: 201. [Tyler Prize for Environmental Achieve-
ment to Willi Dansgaard, Claude Lorius, and Hans
Oeschger; see The Greenland Ice-Core Record section,
this chapter].
Howell, D. G. 1985. Terranes. Sci. Am. 253: 116-125.
Huggett, R. J. 1991. Climate, Earth Processes and Earth His-
tory. Springer-Verlag, Berlin.
International Geosphere-Biosphere Programme. 1992.
Global Change; Reducing Uncertainties. Royal Swe-
dish Academy of Sciences, Stockholm.
Johnson, K. R. 1993. Extinctions at the antipodes. Nature
366: 511-512.
Jones, D. L., A. Cox, P. Coney, and M. Beck. 1982. The
growth of western North America. Sci. Am. 247:
70-84.
Jordan, T. H. and J. B. Minster. 1988, Measuring crustal de-
formation in the American west. Sci. Am. 259: 48-58.
Jouzel, J. 1994. Ice cores north and south. Nature 372:
612-613.
Kanamori, H. and T. H. Heaton. 1996. The wake of a leg-
endary earthquake. Nature 379: 203-204.
Keller, G. 1994. K-T boundary issues. Science 264: 641.
(See also Clifton and Dott, 1994; Kerr, 1994c.)
Kerr, R. A. 1983. Fading El Nino broadening scientists'
view. Science 221: 940-941.
. 1984. Carbon dioxide and the control of ice ages.
Science 223: 1053-1054.
. 1986. Mapping orbital effects on climate. Science
234: 283-284.
. 1987. Milankovitch climate cycles through the
ages. Science 235: 973-974.
. 1990a. Marking the ice ages in coral instead of
mud. Science 248: 31-33.
. 1990b. The ice age bones of contention. Science
248: 32.
. 1990c. Commotion over Caribbean impacts. Sci-
ence 250:1081.
. 1991a. El Nino winners and losers declared. Sci-
ence 251: 1182.
. 1991b. Big squeeze points to a big quake. Science
252: 28.
. 1991c. Yucatan killer impact gaining support. Sci-
ence 252: 377.
. 1991d. The stately cycles of ancient climate. Sci-
ence 252: 1254-1255.
Cause and Effect 83
—. 1991e. Did a volcano help kill off the dinosaurs?
Science 252: 1496-1498.
—. 1991f. Hugh eruption may cool the globe. Science
252: 1780.
—. 1991g. Dancing with death at Unzen volcano. Sci-
ence 253: 271.
—. 1991h. Could the sun be warming the climate? Sci-
ence 254: 652-653.
—. 1991i. Ocean mud pins down a million years of
time. Science 254: 802-803.
—. 1991). Extinction potpourri: killers and victims.
Science 254: 942-943.
—. 1992a. 1991: warmth, chill may follow. Science
255: 281.
—. 1992b. A successful forecast of an El Nino winter.
Science 255: 402.
—. 1992c. Extinctions by a one-two punch?/Extinction
with a whimper. Science 255: 160-161.
—. 1992d. Unmasking a shifty climate system. Science
255:1508-1510.
—. 1992e. Another impact extinction? Science 256:
1280.
—. 1992f. Hugh impact tied to mass extinction. Sci-
ence 257: 878-880.
—. 1992g. A revisionist timetable for the ice ages. Sci-
ence 258: 220-221.
—. 1993a. Pinatubo global cooling on target. Science
259: 594.
—. 1993b. Galileo reveals a badly battered Ida. Science
262: 33.
—. 1993c. Second crater points to killer comets. Sci-
ence 259: 1543.
—. 1993d. Volcanoes may warm locally while cooling
globally. Science 260:1232.
—. 1993e. No way to cool the ultimate greenhouse.
Science 262: 648.
—. 1993f. New crater age undercuts killer comets. Sci-
ence 262: 659.
—. 1993g. The whole world had a case of the Ice Age
shivers. Science 262:1972-1973.
—. 1993h. Even warm climates get the shivers. Science
261: 292.
—. 1993i. How ice age climate got the shakes. Science
260: 890-892.
—. 1994a. Milankovitch plays climate in double-time
in the tropics. Science 263: 174-175.
—. 1994b. Did Pinatubo send climate-warming gases
into a dither? Science 263:1562.
—. 1994c. Testing an ancient impact's punch. Science
263: 1371-1372. (See also Clifton and Dott, 1994;
Keller, 1994.)
—. 1994d. Did the tropical Pacific drive the world's
warming? Science 266: 544-545.
—. 1994e. Baring the secrets of asteroid Ida. Science
266: 1322.
—. 1995a. Darker clouds promise brighter future for
climate models. Science 267: 454.
—. 1995b. How far did the west wander? Science 268:
635-637.
—. 1995c. U.S. Climate tilts toward the greenhouse.
Science 268: 363-364.
—. 1995d. Chilly ice-age tropics could signal climate
sensitivity. Science 267: 961.
—. 1995e. Exonerating an ice sheet. Science 267: 27.
84 Late Cretaceous and Cenozoic History of North American Vegetation

. 1995f. The seasons of global change. Science 267: Michael, A. 1992. Three's a crowd in California. Nature
27-28. 357: 111-112.
. 1995g. Studies say-tentatively-that greenhouse Muller, R. A. and G. J. MacDonald. 1995. Glacial cycles and
warming is here. Science 268:1567-1568. orbital inclination. Nature 377:107-108 [discusses the
. 1996a. Millennial climate oscillation spied. Sci- possibility that "[T]he 100-kyr glacial cycle is caused,
ence 271: 146-147. not by eccentricity, but by a previously ignored para-
. 1996b. Volcano-ice age link discounted. Science meter: the orbital inclination, i, the tilt of the Earth's
272:817. orbital plane."].
. 1996c. Ice bubbles confirm big chill. Science 272: Nadis, S. 1995. Asteroid hazards stir up defense debate.
1584-1585. Nature 375: 174.
. 1996d. A shocking view of the Permo-Triassic. Sci- Newton, P 1995. Perspectives past and present. Nature
ence 274: 1080. 378: 12 [discusses importance of the terrestrial carbon
. 1996e. New way to read the record suggests abrupt sink].
extinction. Science 274: 1303-1304. Oreskes, R, K. Shrader-Frechette, and K. Belitz. 1994. Ver-
. 1997a. Cores document ancient catastrophe. Sci- ification, validation, and confirmation of numerical
ence 275: 1265. models in the Earth sciences. Science 263: 641-646.
. 1997b. Why the West stands tall. Science 275: Post, W M. et al. (+ 5 authors). 1990. The global carbon
1564-1565. cycle. Am. Sci. 78: 310-326.
. 1997c. Second clock supports orbital pacing of the Prance, G. T. (ed.). 1986. Tropical Rain Forests and the
Ice Ages. Science 276: 680-681. World Atmosphere. Westview Press, Boulder, CO.
. 1997d. Moderating a tropical debate. Science 276: and T. E. Lovejoy (eds.). 1985. Key Environments—
1793. Amazonia. Pergamon Press, New York.
. 1997e. More signs of a far-traveled west. Science Rampino, M. R. and S. Self. 1984. The atmospheric effects
277: 1608. of El Chichon. Sci. Am. 250: 48-57.
Krumenaker, L. 1994. In ancient climate, orbital chaos? Robock, A. 1996. Stratospheric control of climate. Science
Science 263: 323. 272: 972-973.
Kunzig, R. 1989. Ice cycles. Discover May: 74-79. Ropelewski, C. F 1992. Predicting El Nino events. Nature
Lacis, A. A. and B. E. Carlson. 1992. Keeping the sun in 356:476-477.
proportion. Nature 360: 297. Ruddiman, W. F and J. E. Kutzbach. 1991. Plateau uplift
Lean, J. and D. Rind. 1996. The sun and climate. Conse- and climatic change. Sci. Am. 264: 66-75.
quences 2: 27-36. Sandweiss, D. H., J. B. Richardson III, E. J. Reitz, H. B.
Lehman, S. 1993. Ice sheets, wayward winds and sea Rollins, and K. A. Maasch. 1996. Geoarcheological ev-
change. Nature 365: 108-110. idence from Peru for a 5000 years B.P. onset of El Nino.
. 1997. Sudden end of an interglacial. Nature 390: Science 273: 1531-1533.
117-119. Schneider, S. H. 1987. Climate modeling. Sci. Am. 256:
Lehodey, E, M. Bertignac, J. Hampton, A. Lewis, and J. Pi- 72-80.
cant. 1997. El Nino southern oscillation and tuna in Searle, M. 1995. The rise and fall of Tibet. Nature 374:
the western Pacific. Nature 389: 715-718. 17-18.
Lewis, J. S. 1995. Rain of Iron and Ice: The Very Real Threat Simkin, T. 1994. Distant effects of volcanism—how big
of Comet and Asteroid Bombardment. Addison- and how often? Science 264: 913-914.
Wesley, Reading, MA. Smit, J. 1997. The big splash. Nature 390: 340-341.
Lorius, C., J. Jouzel, D. Raynaud, J. Hansen, and H. Le Treut. Stanley, S. M. 1984. Mass extinctions in the ocean. Sci.
1990. The ice-core record: climate sensitivity and fu- Am. 250: 64-72.
ture greenhouse warming. Nature 347: 139-145. Stauffer, B. 1993. The Greenland ice core project. Science
Macdonald, K. C. and P. J. Fox. 1990. The mid-ocean ridge. 260: 1766-1767.
Sci. Am. 262: 72-79. Steel, D. 1991. Our asteroid-pelted planet. Nature 354:
and B. R Luyendyk. 1981. The crest of the east Pa- 265-267.
cific rise. Sci. Am. 244: 100-116. Stoffler, D. and P. Claeys. 1997. Earth rocked by combina-
Mahlam, J. D. 1997. Uncertainties in projections of human- tion punch. Nature 388: 331-332.
caused climate warming. Science 278: 1416-1417. Stone, R. 1995. If the mercury soars, so may health haz-
Marvin, U. B. 1973. Continental Drift, the Evolution of a ards. Science 267: 957-958.
Concept. Smithsonian Institution Press, Washington, Sullivan, W. 1974. Continents in Motion, the New Earth
D.C. Debate. McGraw-Hill, New York.
Masood, E. 1997a. Climate panel forecasts way ahead. Na- Tibbetts, J. 1996. Farming and fishing in the wake of El
ture 385: 7. Nino. BioScience 46: 566-569.
. 1997b. El Nino forecast fails to convince sceptics. Walker, G. 1993a. Breaking up is hard to do. Nature 363:
Nature 388: 108. 584-585.
McPhaden, M. J. 1994. The eleven-year El Nino? Nature . 1993b. Back to the future. Nature 362:110.
370: 326-327. Weaver, A. J. 1993. The oceans and global warming. Nature
Melosh, H. J. 1993. Tunguska comes down to Earth. Nature 364: 192-193.
361: 14-15. . 1995. Driving the ocean conveyor. Nature 378:
. 1995. Around and around we go. Nature 376: 135-136.
386-387 [reviews estimates of the size of the Chicxu- Webb, R. S. and J. T. Overpeck. 1993. Carbon reserves re-
lub impact crater]. leased? Nature 361: 497-498.

White, J. W. C. 1993. Don't touch that dial. Nature 364:186.
White, R. S. and D. P. McKenzie. 1989. Volcanism at rifts.
Sci. Am. 261: 44-55.
Whitehead, J. A. 1989. Giant ocean cataracts. Sci. Am. 260:
50-57.
Williams, N. 1997. El Nino slows greenhouse buildup? Sci-
ence 278: 802.
Williams, S. N. 1995. Erupting neighbors—at last. Science
267: 340-341.
Zahn, R. 1994. Core correlations. Nature 371: 289-290.
Zahnle, K. 1996. Leaving no stone unburned. Nature 383:
674-675.
Zuber, M. T. 1994. Folding a jelly sandwich. Nature 371:
650.
Notes
1. For purposes of mathematical manipulation in deter-
mistic modeling, climate is defined as the statistics of one
or more components of the climatic system over a speci-
fied domain and specified time interval, including the
mean, variance, and other moments (Gates, 1982; NRG
Panel on Climatic Variation, U.S. Committee for GARP,
1975).
2. For a discussion of the role of fluctuations in warm,
highly saline marginal seas in driving the thermohaline
circulation of oceans in more ancient times, see Brass et al.
(1982).
3. For an interesting article on atmospheric tempera-
Cause and Effect 85
tures in the Precambrian, the roles of CO2 and weathering
in reducing these temperatures, and the effects on the evo-
lution of ancient biological systems, see Overbeck and
Mancinelli (1994).
4. For a satellite photograph showing the principal
physical features of North America, see Flora of North
America Editorial Committee (1993, fig. 1.5).
5. The fastest present rate of subduction is along the
Tonga Trench at 24 cm/year (Bevis et al., 1995).
6. The terminology used to described objects in space is
complex and is often used inconsistently and without clear
definition outside the field of astronomy. The suffix "-oid"
designates an object while it is in space. A meteoroid is of
silicate or metalic (Fe, Ni) composition, ranging in size
from sand particles to less than 1 km, originating in the as-
teriod belt between Mars and Jupiter, and usually de-
stroyed as it passes through Earth's atmosphere. When a
meteoroid enters a planet's atmosphere it is designated a
meteor, and if it is large enough it can strike Earth's surface
and become a meteorite. A comet is a low-density body
composed mostly of ice around a meteoroid nucleus, orig-
inating in the Oort cloud at the orbit of Neptune and ex-
tending out 100,000 AU (astronomical unites, the mean
distance between the Earth and Sun; 149,600,000 km,
93,000,000 mi), and destroyed gradually by evaporation as
it approaches a sun or a planet's atmosphere. A bolide is
the brightest of the meteors (a fireball or firery meteor). An
asteroid is a big meteor (or small planet) ranging in size
from -1000 km to 1 km in diameter.
7. Bretz's first name is the letter J; it is not an abbrevia-
tion, and hence there is no period.
 THREE
Context
The interaction between vegetation and the environment
over time is one of the most complex of the Earth's inte-
grated systems. In addition to the direct methodologies of
paleopalynology and paleobotany, there are other tech-
niques that provide independent sources of information
for interpreting this interaction. These include paleotem-
perature analysis, sea-level changes, and faunal history.
The first two are also forcing mechanisms as discussed in
Chapter 2, but for this survey the summary curves can also
serve as convenient context information. Each is a vast
subject with an extensive literature, and all are presently
generating considerable discussion. For paleotemperature
analysis, unsettled issues include the extent of temperature
change in equatorial waters during the Early and Middle
Tertiary, which would affect the poleward transport of heat
by conveyer-belt mechanisms. Estimates range from sur-
face waters as warm or warmer than the present to consid-
erably cooler. For the Neogene, CLIMAP estimates based on
the ecology of coccolithophores, diatoms, radiolarians, and
especially foraminifera are that temperatures in the tropics
did not cool significantly; modeling results, terrestrial pa-
leontological evidence, and new Barbados coral data sug-
gest they cooled by ~5°C. There is uncertainty as to when
glaciations began on Antarctica; recent estimates range
from the Early Eocene to late Middle Eocene to Middle
Oligocene (45-35 Ma; Birkenmajer, 1990; Leg 119 Ship-
board Scientific Party, 1988). This affects interpretation of
18
O values during the Paleogene because they could reflect
temperature alone or could be due to ocean water temper-
ature and ice volume changes. Another challenge is to un-
ravel the extent to which benthic temperature records track
insolation-induced changes in water temperature versus
new thresholds in ocean bottom-water circulation.
Discussions of sea-level fluctuation are presently fo-
cused on their causes during the preglacial Early Cenozoic.
In faunal history the timing of the North American Land
86
Mammal Ages (NALMAs) or provincial ages are being re-
vised. For vegetational history much of the older literature
describes events in terms of geofloras, but this conceptual
context, at minimum, requires substantial renovation, and
the boreotropical hypothesis is emerging as an alternative
for envisioning biotic events in the high northern latitudes.
Even with these uncertainties, however, a hallmark of re-
cent studies is the gradual emergence of consistency be-
tween biotic and environmental histories based on the var-
ious independent lines of inquiry.
GLOBAL MARINE PALEOTEMPERATURE CURVE
The history of the Earth's paleotemperature is provided by
analyses of 16O and its isotope 18O in the mineral walls of
marine invertebrates (Savin, 1977, Savin and Douglas,
1985; Savin and Woodruff, 1990). The calcium carbonate-
water oxygen isotope geothermometer is the most widely
applied quantitative tool for estimating ancient ocean tem-
peratures (Savin, 1982).
As noted in Chapter 2, the technique is based on the
theoretical considerations and early demonstration by
Urey (1947), Urey et al. (1951), Epstein et al. (1951), and
Emiliani (1955) that the ratio of 18O and 16O incorporated
into the calcium carbonate walls of such organisms as Coc-
colithophorae, foraminifera, and molluscs is temperature
dependent: more 18O is taken up as the water cools. The
early equations of Urey have been modified several times;
one presented by Shackleton (1974) is
T (°C) = 16.9 - 4.38(5C - 8W) + 0.10(8C - 8W)2,
where T (°C) is the temperature in degrees Celsius, Sc is the
18Q/16Q rat|0 Of tke carbonate wall material relative to a
laboratory standard,1 and 8wis the 18O value of the seawater

in which the wall material was formed. The values are ex-
pressed as deviations from the standard (e.g., -2% = 2
parts/mil less than the standard). It has now been deter-
mined that each 0.23% change in the 18O/16O ratio is equal
to a temperature change of ~1°C. The 18O/16O ratio can be
measured to ~0.I/mil2, which corresponds to a precision of
~0.5°C. During ice-free times, analyses made on benthic
foraminifera record ocean bottom-water temperatures,
which are near freezing and do not change appreciably dur-
ing glacial-interglacial cycles; those made on planktonic
species record the more variable surface temperatures.
DSDP and ODP cores have been studied from the Atlantic,
Indian, and Pacific oceans, and from the northern and
southern hemispheres, including the equatorial regions.
The oldest widespread ocean sediments for which marine
paleotemperatures can be calculated theoretically is -100
Ma; beyond that time most of the older ocean floor has
been subducted. In practice, modification of the wall cal-
cite through time mostly limits the technique to material
-60 m.y. old and younger.
Accurate paleotemperature analysis requires that the
calcium carbonate in the walls of the fossils not be altered
by diagenesis (chemical or physical changes in the sedi-
ment during and after deposition but prior to consolida-
tion). The various kinds of diagenesis that can change the
primary chemistry of the wall include recrystallization;
chemical alteration; dissolution as the shells sink to great
depths; and encrustation, which may overprint or erase the
original isotopic composition. The first three can usually
be detected by careful visual inspection, and the last is ev-
ident through scanning electron microscopy (SEM) or
cathodoluminescence microscopy. The presence of algal
symbionts in some species of planktonic foraminifera may
also cause a problem. As the biology of modern forms be-
comes better known, it is possible to avoid or deemphasize
these species or to compensate for the 18O/16O imbalance
caused by the algae.
Technically, it is now a comparatively straightforward
procedure to obtain 18O/16O data from marine sediments,
and several paleotemperature curves have been published
for the Late Cretaceous and Tertiary (e.g., Miller et al.,
1987; Prentice and Matthews, 1988; Savin, 1977). It is
much more complicated, however, to assign temperature
values to these ratios. These values can be calculated di-
rectly only if the proportion of 18O and 16O in the water has
remained constant over time and if the species analyzed
faithfully accumulate 18O and 16O in equilibrium with the
surrounding water. If some factor alters the relative
amounts of 18O and 16O in the source water from which the
calcium carbonate walls are formed, it is more difficult to
relate a particular ratio to a specific temperature.
Oxygen isotope ratios have changed over time because
water containing the lighter 16O is evaporated and trans-
ported more readily from the ocean surface than the heav-
ier 18O. During interglacial times only -2% of the world's
water is stored in ice and the 16O is returned to the ocean
Context 87
by precipitation (-80% falls on the ocean surface), runoff,
and groundwater seepage. A relatively constant balance is
maintained in the marine environment through the circu-
lation of ocean currents. During glacial times, however, the
return of 16O was delayed because much of the 16O rich
water was held in continental ice sheets (-10%), and the
relative concentration of the heavier (higher atomic
weight) 18O in marine waters increased. At the maximum
of the Pleistocene glaciations (-18 Kya), the mean 18O
value of seawater was +1.2—1.6/mil, during past ice-free
times it was -1.08/mil, and at present it is -0.18/mil
(Savin and Woodruff, 1990). The ratios are a result of both
water temperature and ice volume changes, and at the last
glacial-interglacial interval the amplitude was -1.7%. Ap-
proximately 0.4% was due to cooling and -1.3% was ice-
volume effect (Shackleton and Duplessy, 1986). As long as
there is consensus about the presence of glaciers during a
particular period, the imbalance can be factored into the
calculations to determine temperature. If there is disagree-
ment about the presence of ice, and there is for the Paleo-
gene, different temperature values will be read from the
same ratios.
Disentangling the effects of these two independent vari-
ables is one of the challenges in paleotemperature re-
search. For example, it was assumed previously that in the
Late Cretaceous and until about the Middle Miocene (-14
Ma) in the Tertiary, the Earth was essentially ice free and
that the calcium carbonate of marine waters was in near
isotopic equilibrium with that of the wall material (Shack-
leton and Kennett, 1975a,b). Thus, changes in the ratio
were attributed solely to fluctuating temperatures. For
equatorial surface waters, the present ~28°C temperature
was estimated to have been ~18°C during the Paleogene.
However, recent evidence indicates that mountain glaciers
may have been present in Antarctica in the Early Eocene
(Birkenmajer, 1990), that limited ice sheets were present
along the margin of Antarctica during the late Middle
Eocene and Early Oligocene (Leg 119 Shipboard Scientific
Party, 1988), and that more extensive ice developed later in
the Oligocene (-30 Ma; Miller et al., 1987; Poore and
Matthews, 1984). After that time, development of the
Antarctic ice cap was well under way, and after the Middle
Pliocene (-3.4 Ma) northern hemisphere glaciations be-
came extensive (Shackleton and Opdyke, 1977). For the Pa-
leogene this means that isotopic values of equatorial waters
were due in part to changes in the 18O/16O equilibrium and
that temperatures were warmer than was thought earlier.
Several methods can be used to compensate for the im-
balance in oxygen isotopic equilibrium during glacial cli-
mates. This imbalance, and those derived from local to re-
gional variations in evaporation and precipitation, are
most pronounced near continental margins due to fluctua-
tions in the amount of runoff from streams and glacial
meltwaters. Therefore, one procedure is to base calcula-
tions on planktic foraminifera from the open oceans at low
latitudes where they are less influenced by high-latitude
88 Late Cretaceous and Cenozoic History of North American Vegetation
temperature changes associated with glaciation. A second
method is to adjust the isotope data from the fossils accord-
ing to the temperature requirements of morphologically
similar or identical species. Modern studies recognize that
certain species of foraminifera (e.g., Globigerinoides sac-
culifer) tend to deposit 18O/16O in their walls in close equi-
librium with seawater (Shackleton and Kennett, 1975a),
and these species can be selected or emphasized in the
analyses. This modern analog method provides reliable
data for those parts of the column where several members
of an assemblage are morphologically similar to extant
species, as is frequently the case from the Middle Tertiary
through the Quaternary. Similarities between modern and
fossil forms diminish in the Middle and Early Tertiary, just
when the unsettled timing of continental glaciations also
becomes a problem. The third method is the possibility of
determining the 18O/16O ratios for periods just before, dur-
ing, and immediately after glacial periods and applying a
linear regression2 to remove these trends from the data sub-
sets. From these and other approaches it is estimated that
each 10-m lowering of glacial sea level is equivalent to a
change of -0.1% in 18O (Moore et al., 1982). Because sea-
level fluctuations can be detected in sediments deposited
along the passive margins of continents (see following sec-
tion), these data can be incorporated into estimates of pa-
leotemperatures from 18O values. As noted previously, one
remaining concern is how to distinguish between changes
in marine temperatures due to climatic forcing versus sud-
den threshold reorganization of ocean circulation.
New methodologies are being developed that will even-
tually provide additional data on paleotemperatures, as well
as independent checks on values calculated from 18O/16O ra-
tios. One is based on an index (Uk37) derived from the ratio
of diunsaturated to triunsaturated forms of long chain
(C37-C39) alkenones presently known from sediments ex-
tending back 650,000 years. The index varies with the tem-
perature of the ambient water (as the water cools there is a
higher proportion of the diunsaturated molecule), and it has
been calibrated to specific temperatures through laboratory
culture of the coccolithophorid alga Emiliani huxleyi (Eglin-
ton et al., 1992). A shift in the index of 0.1 equals a change in
temperature of ~3 ± 0.8°C. The technique allows the calcula-
tion of recent paleotemperature changes on a much finer
scale (100 years) than previously possible.
Another ratio is the 13C/12C in organic carbon from ma-
rine sediments. The 13C/12C ratios in organic C35 hopanes
and C27 steranes are preserved in sulfur-bound compounds
likely derived from bacterial plankton at the base of the
photic zone (upper 100 m) in coastal waters (Schoell et al.,
1994). In Miocene sediments from the Monterrey Forma-
tion at Naples Beach near Los Angeles, California, 13C
closely tracked 18O signals in foraminiferal inorganic car-
bonate as the ocean waters cooled.
Another innovation is the use of calcium carbonate
otoliths (ear bones) from fish, which are preserved in the
fossil record. The calcium carbonate is laid down in thin
daily increments and incorporates the oxygen isotope ratio
of the surrounding water. By counting and analyzing these
otolitic equivalents to growth rings, there is the potential
for determining ocean temperature changes at closely
spaced intervals. Isotope fluctuations are already evident
in otoliths from 3.5 Ma in Florida and Idaho, and otoliths
preserved in rocks elsewhere that are as old as 150 m.y.
offer the possibility of extending the record still further.
It is also possible to check results derived from oxygen
isotope measurements from calcium carbonate with mea-
surements of strontium in the carbonate. Strontium can re-
place calcium because it is chemically similar, and the rate
of replacement varies with water temperature. The earlier
CLIMAP estimate of tropical ocean-surface temperature
during the recent glacial maximum (18 Kya) suggested lit-
tle or no cooling (2°C or less) while high-latitude tempera-
tures cooled by ~5°C. This was inconsistent with interpre-
tations made from many terrestrial assemblages and with
modeling, which predicted a cooling at the low latitudes.
Recent measurements of strontium in the carbonate from
Barbados corals suggest that equatorial waters cooled by
~5°C (Guilderson et al., 1994). Further, the oxygen isotopic
composition of seawater sampled during Leg 154 of the
ODP (site 925, Ceara Rise, tropical Atlantic Ocean) suggests
deep water was cooler by 4°C at the last glacial maximum
(Schrag et al., 1996).
A general global marine paleotemperature curve for the
Late Cretaceous and Cenozoic is presented in Fig. 3.1. The
current temperature scale, based on 18O values when ice is
present, is given along the inner left margin of Fig. 3.1; es-
timated values for the ice-free Tertiary are given along the
outer left margin.
The curve shows that during the Maastrichtian ~70 Ma,
ocean deep-water temperatures were between 15 and 20°C,
or about 12-15°C warmer than at present, and gradually
declined to about 10-12°C warmer by the K-T boundary.
Present benthic temperatures average about 1-2°C. In the
debate as to whether climatic change or an asteroid impact
was decisive in the extinction of the large dinosaurs 65 Ma,
dinosaur diversity in the High Plains region did not de-
cline in concert with the gradual lowering of temperatures
shown in Fig. 3.1 but instead declined suddenly and more
rapidly (Sheehan et al., 1991).
Temperatures began to rise during the Paleocene, and by
the Early Eocene values were at or near the highest for all
of Phanerozoic time. The estimated range is between 11
and 15°C for bottom-water temperatures and possibly as
high as 21°C for surface waters at 52° S (Miller et al., 1987).
The present ocean-surface temperatures at the equator are
between about 20 and 28°C. The warming trend across the
Paleocene-Eocene boundary is also preserved in patterns
of paleosol carbonates (a sharp decrease in 13C/12C ratios)
and in mammalian tooth enamel (Koch et al., 1992). The
cause(s) for the amounts of carbon in the biosphere to vary
with temperature is not presently known (P. L. Koch, per-
sonal communication, 1996). A punctuated decline fol-
 Context 89

a. OPENING NE NORTH ATLANTIC GREENLAND/ NORWAY (55Ma)

SEPARATION AUSTRALIA/ANTARCTICA

b. MODERN TRADE WINDS (• 45 Ma)

c. RAPID DROP OCEAN WATER TEMPERATURES (BY - 4 - 5° C).

FORMATION OF COLD BOTTOM WATERS CHARACTERISTIC OF
MODERN OCEANS (- 35 Ma), EARLY ANTARCTIC GLACIERS (33Ma)

d. CLOSING E. MEDITERRANEAN, DISAPPEARANCE OF TETHYS SEA

(BY-21.5 Ma)

e. MESSINIAN SALINITY CRISIS (MEDITERRANIAN RUNS DRY, -4.8- 4.9 Ma)

f. CLOSING ISTHMUS OF PANAMA (- 2.4 Ma) - WARM SALINE WATER

TRANSPORT TO N. ATLANTIC; GULF STREAM INTENSIFIES

g. ONSET MID - LATITUDE N. HEMISPHERE GLACIATIONS (3.0 - 2.4 Ma)

h. PRINCIPAL UPLIFT TIBENTAN PLATEAU (13-7 Ma)

Figure 3.1. Composite benthic foraminiferal oxygen isotope record for Atlantic DSDP sites (based on Miller et al., 1987);
major biological and geological events and selected North American fossil floras are superimposed. Separation of Aus-
tralia from Antarctica, as expressed by shallow water circulation, is placed in the early Eocene (50-55 Ma), and deep
water circulation is evident in the Early to Middle Oligocene (25-30 Ma). The placement of fossil floras along the curve
provides a global paleotemperature context for assessing the paleoenvironments of the assemblages. Data from various
sources.

lowed between the Middle Eocene (52-40 Ma) and the
Late Eocene (40-38 Ma), to 8-10°C warmer. The difference
between the Early to Middle Eocene high and Late Oli-
gocene lows was ~10-11°C. Kennett and Shackleton
(1976) proposed a cooling of 5°C during a period of about
100 Ky. Collectively, the 18O values suggest a mostly ice-
free world between the Late Cretaceous and the Early
Oligocene. Between the Early Oligocene and the Middle
Miocene temperatures fluctuated, but values at the begin-
ning and end of this time span were about the same.
Within this interval the circumpolar Antarctic current de-
veloped, the Antarctic ice sheet formed by the beginning of
the Oligocene (35-30 Ma; there is a significant drop in sea
level at -30 Ma), and there was intensification of cold bot-
tom-water flow. Glaciers likely appeared and disappeared
during the Late Oligocene and Early Miocene (Miller et al.,
1987). A second major decline in temperature occurred in
the Middle Miocene (15-14 Ma). This decline marked the
beginning of permanent ice cover on Antarctica and the
initiation of glacial climates in the Arctic. Temperature gra-
dients increased from low (equatorial) to high (polar) lati-
tudes, and this may have initiated a more vigorous thermo-
haline circulation. A third abrupt lowering of temperatures
is recorded between ~3 and 2.4 Ma, ushering in northern
hemisphere glaciations. These glaciations were well un-
derway by 1.6 Ma, a date used to mark the beginning of the
Pleistocene Epoch, or Ice Age, which extends to the end of
the last glacial advance at -11 Kya.
For vegetational history, the critical question is to what
extent the pattern of paleotemperature change evident in
the marine environment applies to terrestrial habitats. Dorf
(1964) attempted to reconstruct terrestrial paleoclimates
for the Late Cretaceous and Tertiary of the western United
States. The resulting curve (Fig. 3.2) did reflect some trends
 WESTERN EUROPE WESTERN UNITED STATES
Figure 3.2. An early attempt at inferring climatic regimes for western Europe and the western United States from paleobotanical evidence. Reprinted from Dorf
(1964) with the permission of Interscience Publishers.
 Context 91

Figure 3.3. LMA curves from Tertiary floras of the northwestern United States and Alaska. Recent data place the peak of
maximum warmth in the Late Paleocene and Early to Middle Eocene and the sharp drop in the Late Eocene. Adapted
from Wolfe and Hopkins (1967); numbers along the curve refer to fossil floras mentioned in that publication.

subsequently identified from the marine environment. The
warm conditions of the Paleogene, followed by the decline
during the Eocene, and the climatic deterioration of Mid-
dle and Late Cenozoic times are broadly evident. Other fea-
tures, such as the abruptness of the Middle Eocene and
Middle Miocene climatic changes, temperature fluctua-
tions in the Oligocene, and the timing of the events are not
comparable. Overall the curve is not very similar to the one
shown in Fig. 3.1. A curve similar to Dorf's (1964) is given
by Tanai and Huzioka (1967) for Tertiary climatic changes
in Japan. The highly generalized nature of these early re-
constructions, based exclusively on the floristic composi-
tion (modern analog method) of comparatively few floras
and mostly in the absence of radiometric dates, creates the
impression that either marine and terrestrial environments
had rather different climatic histories, the paleopalynolog-
ical and paleobotanical record is not sufficiently sensitive
to closely track global climatic changes, or regional condi-
tions so overprint the response by terrestrial biotas to
global changes that patterns derived from paleovegetation
evidence reflect only local climates.
In 1967 Wolfe and Hopkins published an interpretation
of Tertiary paleoclimates for northwestern United States
and Alaska based on LMA. This important study included
a larger number of floras from a more restricted region than
the analyses of Dorf (1964), and several radiometric dates
were available that provided more accurate information on
the age and stratigraphic relationships among the assem-
blages. Their curve is shown in Fig. 3.3. [Other versions are
given in Wolfe (1978) and Wolfe and Poore (1982)]. The
Wolfe and Hopkins (1967) curve better resembles the later
derived marine paleotemperature curve and suggests that
temperature fluctuations evident from the marine environ-
ment reflect terrestrial paleoclimates more closely than in-
dicated by previous studies.
A comparison was made later between marine paleo-
temperatures and terrestrial paleoclimates using pollen and
spore floras from the Eocene of northwestern Europe (Hub-
bard and Boulter, 1983). Through the use of multivariate
statistical analysis, three paleocommunities were identified
from the total assemblage that were sufficiently close to
modern forest analogs to allow estimates of summer maxi-
mum and winter minimum temperatures. There were warm
periods at -51-53 Ma (top of the London Clay and bottom
of the Bracklesham beds), at ~46 Ma (Headon Beds), and for
brief periods during the Oligocene (Bovey Tracey, Mochras,
92 Late Cretaceous and Cenozoic History of North American Vegetation

cating factor, were tracking genuine global climatic change.

Figure 3.4. (A) Computer results for the mass of
CO2 (in 1018 mol) as a function of time using the
land area curve of Barron et al. (1980) and the cor-
rected spreading rate formulations of Southam and
Hay (1977). (B) Computer results for worldwide
mean annual air-surface temperature as a function
of time corresponding to the situation for CO2 de-
picted in (A). Adapted from Berner et al. (1983).
[See Berner (1994) for revised CO2 values.]
Reprinted with the permission of R. A. Berner and
the American Journal of Science.
Terrestrial records from paleosol carbonates and mam-
malian tooth enamel also closely track fluctuations in the
marine carbon isotope record (Koch et al., 1992).
An early indication of how results integrated from inde-
pendent studies, including paleotemperature analysis, pro-
vide an internally consistent model of past environments
was presented by Berner et al. (1983). Using a modification
of the spreading center rates proposed by Southam and
Hay (1977) and land-sea relationships as described by
Barron et al. (1980), they calculated CO2 values for Middle
Cretaceous through Tertiary time (Fig. 3.4A; revised by
Berner, 1994). When continental position, land-sea rela-
tionships, and CO2 concentration were used to model ex-
pected air-surface paleotemperature patterns, a secondary
temperature maximum was predicted by the model at
-55-40 Ma (Fig. 3.4B). When this curve was compared
with those based on LMA (Wolfe and Hopkins, 1967) and
planktonic carbonate values (Savin, 1977), the similarity in
patterns was clearly evident (Fig. 3.5).
Some uncertainty remains as to the relative roles of or-
bitally induced temperature changes (Wolfe, 1971; Wolfe
and Hopkins, 1967; Wolfe and Poore, 1982) and orogenic
factors (Axelrod, 1981; Axelrod and Bailey, 1969) in the
history of Tertiary vegetation in western North America.
Problems in interpretation arise, in part, from the fact that
while climates were changing, orogenic uplift and volcanic
accumulations were simultaneously providing new higher
altitudinal temperate zones, creating arid habitats to the
lee of developing mountains, altering atmospheric circula-
tion, and intensifying the rate of erosion of silicate rocks.
Nonetheless, warm-cool periods are documented in the
marine realm, and it is likely that the development of ter-
restrial vegetation was influenced by similar patterns
(Wing et al., 1991). Within western North America in par-
ticular, the problem is to achieve a balance between the in-
terwoven roles of orogeny and global climatic change. This
makes even more valuable a resource such as marine
paleotemperature analysis, which serves to better define
one component of this multivariate system.
GLOBAL SEA-LEVEL CURVE

Sea levels are presently changing through a combination of

human and other influences. The Gulf of Mexico is en-
and Lundy beds; 33 Ma); the Oligocene was generally croaching on the Mississippi River delta at a rate of ~1
cooler than the preceding Eocene. These results are similar mm/century because of channels, dikes, and canals that re-
to patterns shown on the curve in Fig. 3.1. When details of duce the input of sediments to the delta. In contrast, Hud-
the diagrams based on paleovegetation evidence were fur- son Bay is rising by ~1 cm/century through isostatic re-
ther compared with 18O/16O data from Shackleton and bound following deglaciation.
Boersma (1981), the major trends were clearly recognizable Sea levels have also changed significantly over the past
in both records. Hubbard and Boulter (1983) concluded that 100 m.y, rising during this time between 1 and 2 mm/year.
there is considerable similarity between the marine and ter- Vail and Hardenbol (1979) have estimated that the differ-
restrial plant record and that the Paleogene floras of north- ence between the highest stand of the Late Cretaceous and
western Europe, where Cenozoic orogeny is not a compli- present levels was over 300 m. If the polar ice caps of
 Context 93

Figure 3.5. Plots of worldwide mean annual air-

surface temperature versus time based on predic-
tions from computer modeling (narrow line), com-
pared with estimates from paleobotanical (Wolfe
and Hopkins, 1967) and 18O studies of planktonic
foraminifera and nannofossils of Cretaceous sedi-
ments of the Shatsky Rise (North Pacific) as sum-
marized by Savin (1977). Reprinted from Berner et
al. (1983) with the permission of the American
Journal of Science.

Antarctica (89% of present total continental ice) and
Greenland (10%) were to melt, sea levels would rise by
-70 m. Melting of the Arctic ice sheet would have less ef-
fect because it is anchored in the Arctic Ocean and the re-
sult on water level would be analogous to the melting of
ice in a glass of water. At the latest maximum of the Pleis-
tocene glaciations at 18 Kya, sea level was more than 100 m
lower than at present (Geophysics Study Committee, 1990;
-121 ± 5, Fairbanks, 1989); in 100 years it probably will be
-0.5-1 m higher through a combination of thermal expan-
sion of marine waters and glacial melting.
Changing sea levels have a variety of effects on climate
and terrestrial plant communities (Warrick et al., 1993).
During low stands, portions of the continental shelf are ex-
posed and become vegetated. The Quaternary glaciations
produced an increase of -15% in land surface, and this al-
lowed greater storage of carbon in the phytomass and in
the adjacent sedimentary basins (Prentice and Fung, 1990).
There is also greater erosion of exposed silicate rocks, and
both processes constitute carbon sinks. At the same time
vegetation buried from previous inundations is uncovered
and decays, and CO2 and methane are released. This pro-
cess constitutes a carbon pump. In turn, higher sea levels
cover portions of the continent, slowing erosion and the
drawdown of CO2.
When sea levels are high, low-lying interior regions of
the continent are inundated. Marine waters have a specific
heat of 0.93 cal/g/°C (compared to that of land with
0.60-0.19 cal/g/°C) and release heat more slowly. High sea
levels result in extensive equable maritime climates, while
low levels promote greater continentality or extremes in
annual temperature variation. Model simulations have
documented that in pre-Cretaceous times continental ex-
tent significantly affected climates in the interior (Yemane,
1993). Crowley et al. (1987), using a 2-dimensional energy
balance model (EBM), found that land-sea configuration
limited summer warming in the central portions of the
large Gondwana continent and was important in the glacia-
tions of the Late Paleozoic. An older model even predicted
that the position and extent of the continents was primary
(Barron and Washington, 1984). However, a newer global
environmental and ecological simulation of interactive
systems model (GENESIS) indicates a mean global cooling
of only 0.2°C with Cretaceous geography, whereas a four-
fold increase in atmospheric CO2 raises surface tempera-
ture by 5.5°C. Although insolation and CO2 concentration
94 Late Cretaceous and Cenozoic History of North American Vegetation
remain principal climate forcing mechanisms, seasonality
must be factored in as lowering sea level drained the
epicontinental sea from interior North America after the
Cenomanian.
Albedo (reflectivity) is the ratio of the amount of solar ra-
diation reflected by the Earth to the amount incident upon
it (Thompson and Barron, 1981). Albedo, along with insola-
tion, drives atmospheric and oceanic circulation; it fluctu-
ates with sea level because the reflectivity of various land
surfaces, vegetation, sea, and ice is different. A rise or fall
by ~100 m would produce a change in albedo of 0.01,
which would alter surface temperatures by ~1°C (Geo-
physics Study Committee, 1990). Barron et al. (1980) dis-
cussed the effect of land-sea distribution and eustatic fluc-
tuations on surface albedos. They concluded that the most
important factor in altering albedo over Late Cretaceous and
Tertiary time was sea-level change. During the Quaternary,
insolation patterns were further strongly influenced by ice
sheet expansion (more reflective ice) and contraction (more
absorptive vegetation) through feedback mechanisms. Al-
bedo values for various surfaces are shown in Table 3.1.
During periods when inundation is extensive, as it was
in the Late Cretaceous, terrestrial biotas may be separated
into geographically and reproductively isolated popula-
tions, which promotes speciation by vicariance mecha-
nisms. A lowering of sea level removes the barrier and pro-
motes speciation through hybridization. These events are
reflected in the fossil record in the form of floristic
provinces that appear and disappear over time (Chapter 5).
It is also noteworthy that 12 of the 13 NALMA boundaries,
established in large part on the basis of immigrant taxa, co-
incide with drops in sea level (Opdyke, 1990; see following
section on Faunas).
The record of sea-level change is preserved as repeating
sequences of near-shore marine deposits (high stands), al-
ternating with erosional surfaces (sequence boundaries) or
with periods of nondeposition identifiable when plank-
tonic foraminifera zones can be tied to a geochronometric
scale (low stands). If no additional constraining geological
information is available, the fluctuations are termed rela-
tive sea-level (RSL) changes because they could be the re-
sult of uplift and subsidence of the land, lowering or rising
of the ocean surface, or both. A distinction between these
two events can be made, however, when the sediments are
deposited along the passive margins of continents. Passive
margins are portions of the continental shelf and adjacent
coastal plain that were relatively stable tectonically during
the period under consideration, as, for example, the At-
lantic and Gulf Coasts during the Cenozoic. This is in con-
trast to active margins, as along the west coast of North
America during the Cenozoic. In the case of passive mar-
gins, the sequences of near-shore marine deposition, sepa-
rated by erosional surfaces or periods of nondeposition
(unconformities), can be attributed mostly to actual changes
in the level of the sea. Such fluctuations are termed eu-
static changes.
Table 3.1. Comparative albedo values for ocean and land
surfaces (from Berger, 1982)a.
Annual global average 14
Ocean
low latitudes 4-7
mid-latitudes 4-19
high latitudes 6-50
Great Lakes
minimum (summer) 6
maximum (winter) 55
Land
desert 20-30
grasslands, coniferous and deciduous forests 15
wetlands 10
tropical rain forests 7
snow-covered land 35-82
Antarctic ice cap 85
Pack-ice in water 40-55
"Reflectivity in percent of incident light during noon
The method for constructing global sea-level curves was
developed at Exxon by a group headed by Peter Vail.
Acoustic signals are sent through underlying strata and are
reflected in a pattern determined primarily by the den-
sity-porosity of the various rock layers. Solid rock returns
a signal of amplitude, frequency, waveform, and velocity
that is different from that of unconsolidated erosion sur-
faces. The result is a vertical seismic profile (VSP) that re-
flects periods of marine deposition (submergence) and ero-
sion (emergence) at a given site.
Seismology has long been used in petroleum explo-
ration as a means of detecting depositional settings and
postdepositional histories favorable to the formation and
accumulation of oil. The innovation of the Vail group, in
applying otherwise standard seismic methodology, was to
correlate among the numerous near-shore sedimentary pro-
files available from Exxon's worldwide operations. The
original curves (Vail et al., 1977) were based on profiles
from about 60 different regions. Within this global net-
work, patterns that showed up at the same time along the
passive margins of distant continents were attributed to
global changes in sea level, while individual isolated pat-
terns were attributed to local tectonics. The differentiation
between local and global changes depends on accurate dat-
ing of the hiatuses on each passive margin, which is
achieved through detailed biostratigraphic, magnetostrati-
graphic, and isotope stratigraphic studies (see Temporal
Context Section).
The sea-level curve developed by Vail and coworkers is
shown in Fig. 3.6. The discussions that followed publica-
tion of the curve centered primarily on three points. One
was the cause of the changes. Glaciation is the most obvi-
ous mechanism for controlling eustatic changes in sea
level (called glacioeustatic changes), and this was pro-
Figure 3.6. Global cycles of
relative sea-level change during
the Tertiary. Reprinted from Vail
and Hardenbol (1979) with the
permission of Oceanus.
96 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 3.7. Correlation between (A) 18O-measured cold-water intervals and (B) low-high
stands of sea level. Based on Chappell and Shackleton (1986). Reprinted from the Geo-
physics Study Committee (1990) with the permission of the National Academy Press.

posed by the Vail group as the principal factor in causing
the simultaneous worldwide appearance of erosion sur-
faces along the passive margins of continents. If so, a corre-
lation between sea level and temperature as recorded by
18
O measurements would be expected, and this was the
case (Chappell and Shackleton, 1986; Shackleton, 1987;
Fig. 3.7). However, rapid falls identified near the end of the
Mesozoic and in the Paleogene did not correspond to any
known periods of major glacial advance. Thus, other fac-
tors must have been involved during preglacial times. One
is a holdover from Cretaceous expansion of midocean ridge
systems with an increase in midplate volcanism (Ager,
1981; Hays and Pitman, 1973; Pitman, 1978; Schlanger et
al., 1981). Ridge expansion decreases ocean basin volume
and displaces water onto the continental margin and into
the interior. If the factors of ridge expansion and thermal
expansion of the crust, upper mantle, and ocean water
were oparating alone, they could theoretically affect sea
level by a -175 m (see Harrison, 1990). Until recently it
was not certain that changes in ridge volume covaried with
the pattern of sea-level change closely enough to identify it
as a significant factor. Gurnis (1990) further suggested that
because the rate at which cold lithosphere is returned to
the mantle along subduction zones has generally been ig-
nored in the models, sea levels might actually rise or fall
when increased spreading is considered alone. Improved
modeling now reveals a better correlation between periods
of continental flooding and increased plate velocities (Gur-
nis, 1993).
Other factors include outgassing of juvenile water
(greatest near the beginning of rapid plate movement, as in
the Jurassic and Cretaceous, and diminishing later); recy-
cling of water through the crust and upper mantle (but this
is mostly a balanced system); changes in the volume of
groundwater (also generally balanced); redistribution of
water among the ocean basins (e.g., the drying of the
Mediterranean in the Late Miocene Messinian salinity cri-
sis, 4.8-4.9 Ma; Hsu et al., 1977), with resulting alteration
in ocean salinity; and the thermal contraction and expan-
sion of water with changing paleotemperatures (steric
changes). Sea level rises -100 mm/l°C of temperature in-
crease throughout the uppermost 500 m. If deep ocean wa-
ters were warmer by 10°C, as was the case during the Late
Cretaceous and Early Tertiary, sea levels would be higher
by -10 m from this effect alone (Geophysics Study Com-
mittee, 1990). Recent measurements off the southern Cali-
fornia coast show that the upper 100 m of water has
warmed by 0.8°C in the past 42 years, resulting in a sea-
level rise of 0.9 ± 0.2 mm/year, which is consistent with
coastal tide gauge records (1-3 mm/year; Roemmich,
1992).
Other factors affecting past sea levels have been reviewed
by Savin and Douglas (1985) and Southam and Hay (1981).
Individually these account for fluctuations estimated from a
few meters to negligible. Collectively, they reduce the
amount of change that previously had to be attributed to un-
known factors or to undetected glaciation for eustacy
throughout the Late Cretaceous and Early Cenozoic.
Another concern was with the suddenness of the de-
clines reflected in the jagged profile of the curve. These
precipitous drops do not match the more gradual rates of
decline evident in some Cretaceous and Tertiary deposits
in the western and midcontinent regions of North America.
However, current research on sedimentary processes along
continental margins is beginning to produce models with
smoother curves (Pitman, 1978). Transport of eroded mate-
 Context 97

rial onto the continental shelf during a time of lowering sea

level may be retarded across freshly exposed portions of
the coastal plain until adequate slope has developed. This
would give the impression from seismic profiles that shelf
erosion had started later and, therefore, that the sea-level
fall occurred over a shorter interval. Results from GRIP and
GRIS2 further reveal that climates do switch from glacial to
nonglacial modes abruptly, so that the sudden changes in
sea level from the Late Eocene onward is not so perplexing.
The third concern was with the magnitude of the
changes. The largest fall in all of Tertiary time was in the
Late Oligocene (-30 Ma) when sea-level lowering was esti-
mated at more than 200 m in 1-2 m.y. When it was as-
sumed that no significant glaciers existed before the Mid-
dle Miocene, the fall was anomolous. It is now known that
this period corresponds to an increase in 18O values in ben-
thic and in high southern latitude planktonic foraminifera
and the development of Antarctic glaciers. The earlier esti-
mate has been revised to just over 100 m (see Kerr, 1984).
Other major declines at about 60, 49.5, 40, 22.5, 10.8, 6.6,
and 4.2 Ma are not so clearly associated with glaciation. An
observation now being incorporated into new versions of
the curve (Haq et al., 1987) is that although the Atlantic
and other continental margins are passive in the sense of
not being active plate boundaries, subsidence is taking
place. The U.S. continental margin is sinking at a rate of
-1-2 cm/1 Ky from continued cooling after the opening of
the Atlantic Basin and from the accumulation of sediments
(sediment loading; Watts, 1982). When tectonic subsidence
at passive margins is taken into account, it reduces the
magnitude of sea-level fall by about half for some parts of
the curve. Revised versions of the curve by Haq et al.
(1987) and Moore et al. (1987) show a gradual scaling
down and smoothing out of the curve as new information
becomes available on the interaction between thermal con-
traction along margins upon lithosphere cooling, the ther-
mal contraction and expansion of ocean waters with tem-
perature changes, sediment loading, midocean ridge crest
volume and volcanism, and glaciation (Fig. 3.8). Miller et
al. (1987) provide a summary that is useful as a working
concept while mechanisms and details of the curve are
being refined. They suggest that the principal mechanism
for major eustatic fluctuations during the past 36 m.y. was
ice-volume change and that for the Mesozoic, Paleocene,
and parts of the Eocene the principal mechanism was the
pace and aftermath of global seafloor spreading. In general,
as more data accumulate, the compilations of Haq et al.
(1987) and Vail et al. (1977) are being validated (Miller et
al., 1996).
During the Late Cretaceous and Early Tertiary (Paleo-

Figure 3.8. Cenozoic cycles of long-term and short-term

sea-level change. Reprinted from Haq et al. (1987) with the
permission of Jan Hardenbol, Peter Vail, and the American
Association for the Advancement of Science.
98 Late Cretaceous and Cenozoic History of North American Vegetation

cene, Eocene), large portions of the Atlantic and Gulf tic history. In transgressive units, marine deposits may in-
Coastal Plains were inundated up to the flanks of the Ap- terfinger with fossil-bearing terrestrial sediments; the latter
palachian Mountains. Most of the continental interior, can then be tied to global events in other environments and
from the Gulf of Mexico to the Arctic Ocean, and from the comparisons can be made between the history of various
Appalachian Mountains to the proto-Rocky Mountains, was biotic groups. Synchronous layers of volcanic debris set-
periodically covered by shallow marine waters. This, in it- tling onto marine, freshwater, and terrestrial surfaces can
self, would suggest equable maritime climates for emergent also provide key markers for dating and correlation (tephra-
North America. When considered in conjunction with pa- chronology). These five techniques of isotopic stratigraphy,
leotemperatures for the same period (Fig. 3.1), a useful con- magnetostratigraphy, biostratigraphy, seismic stratigraphy,
text emerges for estimating the terrestrial climate under and tephrachronology establish a valuable temporal con-
which the vegetation grew. The major drop in sea level at text and provide an impressive arsenal of techniques for
-30 Ma was preceded by a paleotemperature decline of organizing and comparing the records of various biological
~5°C between the Late Eocene and earliest Oligocene. The groups with geological-environmental events.
climatic implications of lower sea level would be greater
continentality in interior climates and changes in the
Earth's albedo. Sea levels and paleotemperatures fluctuated FAUNAS
but without a sustained trend between the Late Oligocene
and the Middle Miocene (-30-15 Ma). Another major low- The North American land mammal record is one of the
ering of sea level occurred at -10.8 Ma, preceded by a sig- most extensive and thoroughly studied in the world
nificant drop in paleotemperatures that probably marks the (Kurten and Anderson, 1980; Savage and Russell, 1983;
beginning of an expanded west Antarctic ice sheet and the Webb, 1985, 1989; Webb et al., 1995; Woodburne, 1987).
early development of Arctic polar ice. After that time the Many North American Late Cretaceous and Cenozoic floras
sea-level curve closely tracks orbital-induced climatic occur intermingled with these faunas; because animal
changes (Matthews and Frohlich, 1991; Prentice and groups are mostly dependent upon the plant formations
Matthews, 1991). Sea-level fluctuations that vary in con- and associations for their habitat and food resources, a gen-
cert with the Milankovitch cycles afford another insight eral correlation among the kinds of fauna, principal vegeta-
into climatic change when 18O data are sparse (Ye et al., tion types, and environmental conditions is evident
1993). throughout the geologic record (e.g., Hutchison, 1982, fig.
5). An awareness of the faunal record, therefore, affords an-
other context for interpreting vegetational history.
TEMPORAL CONTEXT Fossil faunas provide an independent source of infor-
mation on paleoenvironments. Westgate and Gee (1990)
Most rocks undergoing erosion contain a form of iron described a Middle Eocene assemblage from the Laredo
called magnetite. This mineral settles out in an orientation Formation in southwest Texas containing the tropical
aligned with the prevailing magnetic poles. As the mag- Asian palm Nipa. The biological affinities of the other
netic poles reverse periodically, the magnetite tracks their plants could not be established, or the taxa had broad eco-
polarity and position at the time of deposition. Past orien- logical and geographic ranges. The associated fauna in-
tations like those of today are black on the charts presented cluded Carcharias (shark), batoids (skates and rays), bony
in Fig. 5.1, and reversed polarities or anomalies are white. fishes (e.g., Tarpon), and Crocodylidae (Fig. 3.9). Although
The intervals during which a normal (N) or reversed (R) paleoenvironmental reconstructions based on plant mate-
polarity dominate are called chrons (C), and these are num- rial alone would have been equivocal, the associated fauna
bered in a vertical sequence to constitute a magnetostratig- strengthened the interpretation that Middle Eocene cli-
raphy of relative ages. The duration of a reversal is gener- mates in the region, and the ecological parameters of the
ally between 1000-6000 years and the most recent was the Eocene nipas, were warm temperate to tropical. Faunas re-
Brunhes-Matuyama reversal at -780,000 years ago. Mag- flecting similar warm environments have been reported
netochrons in continuous deep-sea sections have been tied from Upper Cretaceous and Lower Tertiary deposits in the
to radiometric dates to form a magnetic polarity time scale northern Great Plains (turtles, crocodilians, champsosaurs;
of absolute ages. In some charts, units of the magne- Hutchison, 1982) and from Lower to Middle Eocene de-
to stratigraphy are correlated with intervals designated P posits from Ellesmere Island within the Arctic Circle (alli-
(e.g., Plb at chron C29-28, -63 Ma). These are planktic gatorlike crocodilians, turtles, perissodactyl mammals;
foraminiferal zones. C or NP designates a calcareous or Dawson et al., 1976; McKenna, 1980; West and Dawson,
nannofossil zone and together they allow further compari- 1978). The lower temperature limit of modern Alligator is
son with environmental and biotic events in the marine ~4.4°C. This indicates that warm-temperate to tropical cli-
realm (marine biostratigraphy). If seismic profiles are avail- mates extended far into the high latitudes, at least along
able, these can be used to establish another chronologic maritime coastal areas (including the margins of the epi-
framework for faunal (NALMA) provincial ages and floris- continental sea), during the Late Cretaceous, Paleocene,
 Context 99

Figure 3.9. Habitats of modern relatives of paleoecologically useful species in the Casa
Blanca fauna (Laredo Formation, Middle Eocene, Webb County, Texas). TR, tropical; SU,
subtropical; WT, warm temperate; TE, temperate; T, terrestrial; F, fresh water; E, estuarine;
N, near shore; I, infralittoral; C, circalittoral; O, oceanic. Reprinted from Westgate and Gee
(1990) with the permission of Elsevier Science-NL.

and Early to Middle Eocene; interior portions of the conti-
nents may have been more seasonal. Such reconstructions,
in turn, are consistent with the low physiographic relief of
the period (Chapter 2), widespread epicontinental seas as
evidenced from the global sea-level curve (Figs. 3.6, 3.8),
high CO2 concentrations, and warm conditions reflected by
the oxygen isotope based paleotemperature curve (Fig. 3.1).
In the Late Tertiary and Quaternary cooling climates are re-
flected in the large size attained by certain mammals. Ac-
cording to Bergmann's Rule, body size increases with de-
creasing temperatures. With larger size, volume and the
capacity for heat production both increase more than skin
surface and the capacity for heat loss. This trend in body
size parallels the climatic deterioration after the Middle
Miocene.
Fossil faunas provide an indication of the type and
structure of vegetation prevailing in a region during a par-
ticular span of time. The increasing diversity and abun-
dance of horses, canielids, and other browsing and grazing
ungulates (hooved mammals) in North America beginning
in the Late Eocene parallels the spread of seasonally dry
woodland, savanna, and, eventually, grasslands. These
low-density communities developed under dry conditions,
and the grassland formation characterizes the interior low-
lands where only moderate physiographic relief existed for
much of the Tertiary. The physical conditions under which
these vegetation types grew were not as conducive to the
preservation of extensive and diverse floras as were those
of the temperate deciduous forests. Vegetation surrounding
the numerous lakes formed through volcanic activity in
100 Late Cretaceous and Cenozoic History of North American Vegetation
western North America during the Tertiary (Chapter 2) or
growing along the shores of a gradually retreating epicon-
tinental sea, as with the tropical to warm-temperate Eocene
Mississippi Embayment floras, was more likely to be pre-
served. The presence of faunas typically associated with
drier open forests, savanna, or grasslands can be especially
useful in reconstructing the kind of plant formations and
associations present when direct paleobotanical evidence
is meager. Janis (1984) emphasized the usefulness of the
different kinds of fossil ungulate communities in estimat-
ing the structure of vegetation growing under seasonally
dry or cold climates.
As would be expected with any complex system, the
faunal and plant evidence is not inevitably consistent
through all of Late Cretaceous and Cenozoic time. There
are instances where faunal assemblages and paleobotanical
evidence from the same formation suggest different vegeta-
tion types and paleoenvironments. As mentioned earlier,
such situations have often provided fodder for entertaining
controversies that mostly make for good theater and bad
science. The modern approach is to view these apparent
incongruities as opportunities to refine existing concepts
and to provide a better understanding of past biotic-
environmental interactions.
As an example, Neogene deposits associated with the
Columbia Plateau region preserve an extensive sequence of
floras that contain a rich assemblage of deciduous forest el-
ements now found in eastern North America, western
North America, and temperate Asia. Extant and extinct
species of Acer, Alnus, Betula, Carya, Castanea, Cornus,
Corylus, Diospyros, Engelhardia, Fagus, Fraxinus, Ginkgo,
Glyptostrobus, Ilex, Juglans, Liquidambar, Magnolia,
Nyssa, Ostrya, Platanus, Pterocarya, Populus, Quercus,
Salix, Sassafras, Tilia, and Ulmus were present in a region
that today is mostly covered at low to middle elevations by
an Artemisia (sagebrush) cold desert and riparian vegeta-
tion. The physiognomy of the vegetation suggested by the
modern analogs is a dense hardwood forest. In contrast,
many of the associated fossil mammalian faunas are rich in
browsing and grazing ungulates more typical of open
forests, savannas, and grasslands. The paleoclimatic impli-
cations of a landscape dominated by hardwood forest, ver-
sus one covered by grassland or savanna that supports
herds of grazing mammals, are quite different and require
either a plausible explanation or reinterpretation of one of
the data sets. As it turns out, an explanation is available.
As noted in Chapter 2, one reason for the extensive se-
quence of fossil floras and faunas in western North Amer-
ica was the presence of conditions favorable to fossiliza-
tion created by extensive volcanic activity. The lava flows
blocked drainage systems and created numerous lakes that
served as depositional basins, while ejection of large quan-
tities of ash provided the matrix for exceptional preserva-
tion in fine-grained shales, tuffs (water-deposited ash), and
diatomite. The extent of volcanic activity in the region is
reflected by the sedimentary sequence shown in Figs. 3.10
and 3.11 from the Late Miocene Trout Creek locality in
southeastern Oregon. The light bands in Fig. 3.10 are de-
posits of diatomite, and the grey ones are compressed lay-
ers of ash. Diatomite is a rapidly accumulating sediment,
and many of the ash bands were also probably deposited in
tens to a few hundreds of years. The approximately 20 lay-
ers of ash evident in the 55-ft section are testimony to fre-
quent volcanism.
Taggart and Cross (1990) and Taggart et al. (1982) used
evidence for volcanic activity from the nearby, and slightly
older, Succor Creek flora to develop a model for explaining
the association of abundant browsers and grazers with fos-
sil floras that reflect dense forests. Fine-resolution sam-
pling from zones immediately above the ash layers pro-
vided pollen and spore profiles that revealed successional
stages (seres) of grasses and other herbs that were replaced
by climax forest communities. These intervals were brief,
lasting only a few to several hundred years, and would go
undetected in normal megafossil sampling procedures.
The overall regional plant fossil record would therefore
suggest that forests dominated the landscape during virtu-
ally all of the Miocene. However, frequent volcanic activity
provided a shifting mosaic of open, short-lived communi-
ties that were sufficient to sustain mobile herds of grazing
(Merychippus, Dromomeryx] and woodland-browsing (An-
chitherium, Hypohippus, Ticholeptus) horses and other
ungulates. During times of diminishing food supply fol-
lowing volcanic eruptions, herds would likely congregate
around lake margins, enhancing their representation in the
faunal record of the region (e.g., Shotwell, 1968). The dif-
ferences are resolved in a model that emphasizes short-
term successional, as well as long-term climatic, changes.
Fossil faunas can provide an estimate of the age and
stratigraphic relationships between strata when radiomet-
ric dates from plant-bearing strata are absent or inconclu-
sive. The scale is the NALMA or provincial ages (Opdyke,
1990; Webb, 1985; Fig. 3.12; see Chapter 6 Context Sum-
mary). Each age is characterized by a fauna that can pro-
vide information on the paleoenvironments and expected
associated vegetation types at specific times during the
Late Cretaceous and Cenozoic. The NALMAs can be fur-
ther plotted with reference to the paleotemperature curve
constructed from the marine oxygen-isotope record (Fig.
3.13). It should be noted that ages for the subdivisions of
the Paleogene are being revised (Berggren et al., 1992,
1995; D. R. Prothero, personal communication, 1996), and
some may be older by ~2 m.y. and two polarity chrons (see
Chapter 6). Some selected fossil floras are included in Fig.
3.13 as examples of the context that can be established for
interpreting paleovegetation.
The fossil faunal record can often identify or confirm
pathways of migration and allow more precise reconstruc-
tions of land connections than would be possible from
plant evidence alone. Most plants and many animals can
move across narrow to moderate barriers, but the inter-
change of large mammals more precisely defines the time

and place of complete land connections. The pattern of
new immigrations into North America from Asia via
Beringia, from Europe via the De Geer and Thulean routes,
and from South America through the Isthmus of Panama
are preserved in pulses of affinities with each area. The
timing and paleoenvironmental implications of these ex-
changes were reviewed by Webb (1985).
The Late Cretaceous Lancian NALMA is represented by
the fauna of the Lance Formation in Wyoming and by the
Hell Creek fauna of eastern Montana. The assemblages
consist of Multituberculata (an extinct group of Mesozoic
and Paleogene rodentlike organisms), Marsupialia (marsu-
pials), and Eutheria (placentals). The depositional settings
were swamps and waterways bordering the Cretaceous
epicontinental sea. Primary geographic affinities are with
Asia and Europe; there is no evidence of direct inter-
change with South America after the Maastrichtian and
for most of the Tertiary. This reconstruction is consistent
with the paleophysiography of the North Pacific and North
Atlantic region during this period; connections with
South America did not become established until late in
the Pliocene.
The Paleocene (Puercan, Torrejonian, and Tiffanian
land-mammal ages) witnessed extensive adaptive radiation
of the placentals (12-13 species, double that of the Lan-
cian), the marsupials declined from 13 to two species, and
the dinosaurs disappeared completely (Webb, 1985). Three
orders of large herbivores were present in the Middle to
Late Paleocene: Pantodonta, Taeniodonta, and Dinocerata.
The latter two were browsers, and all three orders probably
inhabited local open woodlands (Webb, 1989). Late Creta-
ceous and Paleocene faunas have been described from sites
ranging from Texas to Alberta, and their composition gen-
Context 101
Figure 3.10. Exposure of the
Trout Creek diatomite (Late
Miocene), southeastern Oregon;
ash layers represent volcanic
eruptions. Adapted from
Graham (1965).
erally reflects widespread tropical to warm-temperate fau-
nas distributed latitudinally along low thermal gradients.
The notable zoogeographic feature of the Wasatchian
(Early Eocene) is the especially close affinity with Euro-
pean faunas. About one-half of the known land-mammal
genera of North America were also present in Europe
(mostly a consequence of migration from North America to
Europe). This represents the strongest mammalian affinity
that ever existed between the continents and suggests the
North Atlantic land bridge was in full operation during the
Early Eocene, both in terms of physical connection and cli-
mates suitable for the interchange of warm-temperate to
subtropical forest-inhabiting mammals. The faunal evi-
dence is, in turn, consistent with geological (McKenna,
1975; Chapter 2) and paleobotanical reconstructions
(Tiffney, 1985) for the region.
In the Late Eocene, similarities between North Ameri-
can and European mammalian faunas declined but were
maintained with Asia, a scenario compatible with the re-
spective geologic history of the two land bridges. Also, a
drop in sea level occurred at -40 Ma in the Late Eocene
(Duchesnian NALMA), which maintained the land con-
nection between Asia and North America. Toward the end
of the Eocene the temperature decline intensified a trend
toward faunas occupying seasonally dry to more arid habi-
tats (e.g., camelids), and Webb (1989) believes "that wood-
land savanna had become the predominant biome in North
America, displacing the subtropical forests that had pre-
vailed for the first 17 million years of Tertiary history." To a
degree this trend is consistent with the composition and
leaf physiognomy evidence from the late Middle Eocene
Green River flora (46 Ma) of Colorado and Utah (MacGini-
tie, 1969) in which Leguminosae, Sapindaceae, Anacar-
102 Late Cretaceous and Cenozoic History of North American Vegetation

diaceae, species with smaller leaves, early grasses, and

evaporitic and oxidized redbed deposits are evident.
In Chron 13R (37 Ma) there was another lowering of sea
level, and at about 31-29 Ma (early Late Oligocene) a major
fall of up to 200 m occurred (Figs. 3.6, 3.8) concomitant
with a moderate decline in global temperatures (Fig. 3.1).
Extensive land connections were present through Beringia.
In the midcontinent region of North America floral diver-
sity decreased, and LMA from the Florissant flora of Col-
orado reveal a decline in entire-margined leaves. Hutchi-
son (1982) found a contemporaneous decline in aquatic
reptile diversity. The faunal history of the Oligocene in Eu-
rope has been called the "Grande Coupure" (the great
turnover) because of differences between the Early and
Middle Tertiary assemblages. This was due primarily to the
decline of immigrants from North America with disruption
of the North Atlantic land bridge. New waves of immi-
grants arrived in North America from Asia through
Beringia during the Chadronian. The introductions re-
sulted in a change of -60% in the North American land-
mammal fauna and included, in particular, rodents (het-
eromyids, geomyids, castorids, sciurids, cylindrodontids,
cricetids) and larger mammals (Canidae, Felidae, Mustel-
idae, Tapiridae, Rhinocerotidae, Anthracotheriidae, Tayas-
suidae). There was a pronounced shift from small mammal
and arboreal forms, including many primates, to increased
representation of terrestrial rodents and large herding un-
gulates of savanna and open-forest habitats (Mesohippus,
Leptomeryx, Poebrotherium, Meyrcoidodon). These faunas
persisted generally throughout the Oligocene into Heming-
fordian time (Miocene; Figs. 3.12, 3.13). New introductions
from Asia slowed in the Barstovian and Clarendonian.
Brown and Heske (1990) present interesting results on
the effect that new introductions and exclusions (emigra-
tion, extinction) of rodents can have on the development of
vegetation types. When three species of kangaroo rats
(Dipodomys) were excluded from a Chihuahuan Desert
shrub habitat in southeastern Arizona, there was a three-
fold increase in the density of tall perennial and annual
grasses—a trend from shrubland to grassland. These data
suggest that the introduction and persistence of new ro-
dent and large mammal populations during the Cenozoic
involved not just a passive range expansion into favorable
vegetation types already determined by physical (land con-
nections) and climatic factors, but that once introduced,
their presence and later absence likely influenced the his-
tory of shrublands, grasslands, and savannas.
Miocene immigrants included cats (Pseudaelurus),
bears (Ursavus, Hemicyon), beaver (Anchitheriomys, Cas-
tor), flying squirrel (Blackia), and the proboscideans (Mz'o-
Figure 3.1 1 . Diagram of
the Trout Creek di- mastodon, Gomphotherium). These mostly reflect a mosaic
atomite section show of forest, open forest, and savanna vegetation across many
ing number and extent parts of the continent. There was an increase in the diver-
of ash layers (stippled). sity of browsing and grazing herbivores, including about
Adapted from Graham 12 genera of horses (three browsers, nine grazers) and
(1988). nearly as many genera of camels. The number of native un-
 Duration
Epochs and Ages (years x 106)

Pleistocene
Rancholabrean 0.5
Irvingtonian 1.3
Pliocene
Blancan 3.2
Miocene
Hemphillian
Clarendonian 4.
3.0
Barstovian 4.00/
Hemingfordian 4.0
40
4.0
4.0 /
Arikareean 4.
Oligocene
Geringean 4.0 /
Whitneyan 20
Orellan 2.0 /
Chadronian 5.0
Eocene
Duchesnean 2.0
Uintan 8.0
Bridgerian 2.0
Wasatchian 4.0
Clarkforkian 2.0
Paleocene
Tiffanian 4.0\
Torrejonian 2.0 \
Puercan 3.0 \
Cretaceous
Lancian 5.0 \

Figure 3.12. NALMAs and magnetic polarity time scale illustrating faunal context with
which fossil floras can be compared. Compiled from Webb (1985) and Opdyke (1990).
Note that recent age revisions (Berggren et al., 1992,1995; Prothero, 1994; D. R. Prothero,
personal communication, 1996; Prothero and Swisher, 1992) for the Paleogene are af-
fecting NALMA provincial ages by ~+2 Ma (Chapter 6). Reprinted with the permission
of S. David Webb, Plenum Publishing Corporation, and The University of Chicago Press
(Journal of Geology), The University of Chicago.
104 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 3.13. Composite Ceno-
zoic 18O record for benthic
foraminifera compared to
NALMA; selected fossil floras
are included as examples of the
faunal context available for in-
terpreting vegetational history.
Adapted from Opdyke 1990.
(See note in caption to Figure
3.12.) Reprinted with the per-
mission of The University of
Chicago Press (Journal of Geol-
ogy), The University of Chicago.
gulate genera doubled about every 3-5 m.y. from the Hem-
ingfordian to the end of the Clarendonian. The reason for
the increase is attributed to vegetational changes from
woodland savanna to grassland savanna in many parts of
North America (Webb, 1989). Paleobotanical evidence for
the Late Eocene and later Cenozoic reflects forest and
woodland vegetation; the faunal record suggests more
open forests, savannas, and grasslands. The Miocene biota
in parts of central North America has been compared to
that of the modern African savannas (Webb, 1977); but as
Janis (1984) points out, there are significant differences be-
tween the vegetation and ungulate communities in the two
regions. Regardless of quantitative considerations, how-
ever, it is clear that the environmental and vegetational his-
tory patterns reflected by the flora and fauna are generally
consistent.
Another major faunal turnover occurred in the Blancan,
and more than two-thirds of the first mammals to appear
were new introductions. This coincided in time with the
Messinian (Mediterranean) salinity crisis, expansion of the
great African savannas, and the divergence there of the
ape-human lineage. The new forms in North America in-
cluded a large number of Asian microtid rodents, Petau-
rista (a large, subtropical flying squirrel), Parailurus (an ex-
tinct relative of the lesser panda), Ursus (brown bear),
Cervus (elk), Homotherium (a hyaenid), and Brachypoth-
erium (a rhinocerotid that later gave rise to the amphibious
North American genus Teleoceras). A distinctive feature of
the new assemblage was an increased contingency from
South America, including the sloth (Glossotherium), ar-
madillos (Dasypus, Holmesina, Glyptotherium), and por-
cupine (Erethizon). This is consistent with estimates for
uplift of the Isthmian region that formed a continuous land
bridge by -3.5 Ma (Fig. 3.14).
In the latest Miocene the trend toward open forests, sa-
vannas, and grasslands continued; but the development of
even colder and drier conditions reduced the rich savanna
ungulate fauna in North America. In the Pleistocene the
formation of Arctic and alpine tundras, coalescence and
spread of gymnosperms and associated elements into the
extensive boreal coniferous forest or taiga, the appearance
of deserts, the pronounced altitudinal zonation of commu-
nities within the western Cordilleras, and severe winters in
the midcontinent region resulted in a provincialization of
the North American fauna and increased differentiation of
regional communities (Webb, 1989). An overall trend dur-
ing the Pleistocene was the replacement of large herbivores
by smaller herbivores (Webb, 1969). The extinction of the
 Context 105

Figure 3.14. Immigrant land-mammal genera in North America during the past 70 m.y.
Reprinted from Webb (1985) with the permission of S. David Webb and Plenum Publish-
ing Corporation. (See note in caption to Figure 3.12.)

megafauna as a result of rapid climatic changes near the
end of the Pleistocene was accelerated by Asian hunters
who had crossed the Bering land bridge into North Amer-
ica (Barnosky, 1989; Martin and Wright, 1967).
It is clear from this brief summary that a knowledge of
faunal history can provide valuable insights and support-
ive data for estimating the kinds of vegetation and environ-
ments existing in North America in the Late Cretaceous
and Cenozoic. It can also help interpolate critical strati-
graphic control including relative and absolute age data.
Finally, the faunal record can provide information relevant
to certain conceptual questions that are important in how
we envision the history of plant communities.
Paleoenvironmental reconstructions for the Cenozoic
are based, in part, on the present occurrence of presumed
modern analogs. Environmental conditions such as rainfall
and temperature regimes and altitudes are compiled, indi-
vidual members of the assemblage are sorted out into
paleocommunities according to the associations and eco-
logical characteristics of their counterparts in the modern
biota, preliminary environmental reconstructions are made,
and the models are compared and refined with reference to
the ancillary sources of data. If the modern ranges of some
elements do not reflect their full ecological amplitude but
reflect only that part into which they have moved since the
last climatic change or barrier removal, this will be a weak
point in the reconstruction. Often the consequences of these
individual cases are mitigated by considering the ecologi-
cal requirements of the entire assemblage; but if the species
is distinctive, conspicuous, and/or numerically abundant,
or if its significance is overemphasized for some reason,
artifically narrow limits may be placed on the assumed
paleoenvironment. Where an organism does occur is not
always the same as where it can occur, and this concept
106 Late Cretaceous and Cenozoic History of North American Vegetation
is important in tracing environmental change and biotic
history.
A case in point is provided by the migratory history of
the edentate Dasypus novemcinctus (nine-banded arma-
dillo; Humphrey, 1974). This is a South American genus
that first appeared in North America in the Late Blancan
mammal age after uplift of the isthmian land bridge -3.5
Ma (Coates et al., 1992; Graham, 1992). It was first reported
in northern Mexico and south Texas in 1854 (climatic data
for Brownsville: January average temperature 59°F, annual
rainfall 32.9 in.; United States Department of Agriculture
Yearbook, 1941). By 1974 it had moved into Colorado,
Kansas, and Missouri (climatic data for St. Louis: January
average temperature 32.9°E annual rainfall 36.67 in.). Al-
though a warming climate is clearly a factor in the contin-
ued expansion northward, removal of barriers and migra-
tion rate (~4-10 km/year) have been important in
determining the range of Dasypus since its introduction
into North America. Paleoenvironmental reconstructions
based on assumed ecological parameters from fossil and
much of the pre-1854 distribution, compared to the pres-
ent, would be different and a result of the fact that the
range of Dasypus did not then reflect its full ecological am-
plitude. The northernmost limit of Alligator presently con-
forms to the 10°C isotherm; but when historical records are
taken into account, the 4.4°C isotherm is a better estimate
(Hutchison, 1982; Markwick, 1994). These and other exam-
ples from the faunal record reinforce emerging data from
the plant record, especially from Quaternary pollen and
spore studies, that many species are capable of living
under a wider range of ecological conditions than their
present distribution implies.
Another conceptual view is that communities are fluid
in composition over time and that the particular versions
witnessed today are only a few among many possible com-
binations. [For a similar view based on insect communi-
ties, see Elias (1991); for mammals, see the FAUNMAP
Working Group (1996)]. The species comprising an associa-
tion at any given point in time is a reflection of ecological
compatibility; the opportunity to associate as determined
by the availability and dispersal-survival potential of
propagules in relation to existing barriers; the presence of
soils suitable for each component; epidemic disease; time;
and evolutionary alteration in the ecological parameters of
individual lineages. The latter is frequently alluded to in
the literature; but widespread fluctuations in the ecological
requirements of a species through Cenozoic time, unac-
companied by any morphological change that would lead
to its recognition as a novel taxon, are easier to assume
than document (Elias, 1991). The fossil record generally re-
flects ecologically (rather than taxonomically) defined as-
sociations that are recognizable through time, but which
frequently do involve the continued association of some of
the dominants. My impression is that the definition and
dynamic nature of plant paleocommunities is due more to
ecological compatibility; the vicissitudes of propagule
availability and dispersal potential; the existence, kinds,
and extent of barriers; and the length of time since the last
environmental reshuffling than to rapid widespread
changes in the ecological requirements of numerous taxa
accompanied by morphological stasis. A similar view is
expresed by Hubbard and Boulter (1983), who note "that in
certain family and generic groupings there has been no
gross overall change in climatic and ecological prefer-
ences." Nonetheless, plant and animal communities are,
within limits, temporal assemblages that undergo changes
in composition through time. A contrasting view places
more emphasis on the movement of plant communities as
recognizable blocks in response to environmental change.
No treatment of the Late Cretaceous and Tertiary history of
North American vegetation would be complete without
some comment on the turbulent history of geofloras.
GEOFLORA CONCEPT
The genesis of this idea was the application of the name
arctotertiary flora by Engler (1879/1882) to an extant vege-
tation "Distinguished by numerous conifers and the nu-
merous genera of [deciduous angiosperm] trees and shrubs
that now dominate in North America or extratropical East-
ern Asia and in Europe" (quoted from Mai, 1991).
Ralph W. Chaney (Museum of Paleontology, University
of California, Berkeley) began his studies on the Tertiary
floras of the western United States early in this century
(e.g., Chaney, 1920). [For a synopsis of his life and work see
Andrews (1980) and Gray and Axelrod (1971)]. By the
1950s he had recognized that Oligocene and Miocene flo-
ras there typically contained a strong temperate deciduous
angiosperm and coniferous element, in contrast to older
floras with a more tropical aspect, and younger ones with a
strong subhumid (dry) component. Chaney added a time
dimension to extant vegetation types with the term ge-
oflora: "a group of plants which has maintained itself with
only minor changes in composition for several epochs or
periods of earth history, during which time its distribution
has been profoundly altered . . ." (1959, p. 12).
Three geofloras were proposed for North America. In
the Paleogene the Neotropical-Tertiary Geoflora extended
into the Mississippi Embayment region, as represented by
the Middle Eocene Claiborne flora. In the west the Clarno
flora preserves an assemblage with tropical elements that
reached northern Oregon. Individual elements extended
even further north, occupying progressively more coastal
sites in maritime environments. The Neotropical-Tertiary
Geoflora reached its maximum northern extent during the
Early Eocene, after which time it became progressively
constricted toward equatorial regions in response to cool-
ing and seasonally dry climates. The present northern limit
of tropical vegetation in the sense of a tropical rain forest in
North America is in southeastern Mexico in southern Ver-
acruz state (17° N), while individual elements such as Rhi-

zophora (red mangrove) extend northward along the
Florida coasts.
The Arcto-Tertiary Geoflora was described as a temper-
ate deciduous angiosperm forest that presently characterizes
much of eastern North America and parts of eastern Asia.
The name implies a place (the Arctic) and a time of origin
(the Tertiary) and, after the temperature decline of the late
Eocene, it was viewed as extending its range progressively
southward. The Arcto-Tertiary Geoflora was also identified
in Early Tertiary high-latitude floras from Asia and Europe.
It was considered to be a nearly continuous band of temper-
ate vegetation encircling the middle latitudes of the north-
ern hemisphere during the Middle Tertiary.
The Madro-Tertiary Geoflora (Axelrod, 1958) is a vege-
tation of dry to arid environments. Although early lineages
and pre-adapted elements were present in the Paleogene,
this geoflora became recognizable as a distinct vegetation
type during the Mio-Pliocene. This was in response to re-
duced rainfall created by uplift of the coastal cordilleras
and the Sierra Nevada and, to the north, by colder climates
(viz., the cold deserts of much of the Basin and Range
Province). From the perspective of geofloras, the history of
North American vegetation was seen as a shifting mosaic of
these three communities over the landscape in response to
climatic and physiographic changes.
Much of the discussion about geofloras has centered on
the Arcto-Tertiary Geoflora. Its early history was confused
by an erroneous age assignment of plant-bearing strata in
the Arctic. The Swiss paleobotanist and entomologist Os-
wald Heer (Andrews, 1980) was familiar with the Euro-
pean Neogene floras that had a strong temperate compo-
nent. When floras from the Arctic region were discovered
with temperate representatives (Heer, 1869), they were as-
signed a similar age; and an "Arctic Miocene" came into ex-
istence with floras that included broad-leaved deciduous
elements. These floras are now known to be Paleocene and
Eocene in age. The origin of the Arcto-Tertiary Geoflora
was pushed back to the Late Cretaceous and Paleogene in
the form of "blurred preliminary prints" (Chaney, 1967).
Although not emphasized in early considerations, these
high-latitude temperate deciduous Paleogene floras are as-
sociated with tropical elements. Rollick (1936) described
an assemblage from Berg Lake, Alaska. He identified Popu-
lus, Juglans, Planera, Ulmus, Rhamnus, Cornus, Rhodo-
dendron, and Fraxinus, as well as Artocarpidium, Mohro-
dendron, Magnolia, Cinnamomum, Persea, Malpoenna,
Terminalia, and Semecarpus. The flora has since been re-
vised to include Dryopteris, Platycarya, Alnus, Knema,
Myristica, Cinnamomophyllum, Luvunga, Malanorrhoea,
Celastrus, Parashorea, and Alangium. LMA has shown
that 71% of Hollick's "species" have entire-margined
leaves (Wolfe, 1977), and a number of paleotropical genera
are now recognized (e.g., Parashorea]. Fan palms are re-
ported from the Paleocene of Alaska (57° and 61.5° N; Hoi-
lick, 1936) and from the Paleocene of Greenland (70° N;
Koch, 1963). Assessment of the geoflora concept partly de-
Context 107
pends on whether the tropical elements are viewed as in-
termingled with temperate ones (based on present-day
habitats of similar taxa) to constitute an assemblage unique
in taxonomy and ecology, mostly separated in coastal and
lowland habitats from inland and upland temperate ele-
ments, or part of a suspect terrane. This is unsettled. The
current trend is toward interpreting the tropical elements
as largely a coastal fringe community that was transported
northward, but only moderate distances, during the Paleo-
gene. This would leave intact a version of a deciduous for-
est formation that was not anomalously mixed extensively
with tropical species. At a minimum, however, modifica-
tion of the Arcto-Tertiary Geoflora would have to include
an associated paratropical rain forest with Old World com-
ponents. Valid criticisms of the original version of the ge-
oflora concept are that it was based on an erroneous stratig-
raphy that affected estimates of its time of origin, presented
too simplistic a view of vegetation types in the Arctic dur-
ing the Paleogene, and underestimated the importance of
Old World tropical components.
The principal concern, however, centers on the claim
that each geoflora has "maintained itself with only minor
changes in composition for several epochs or periods of
earth history." The extent to which Chaney had come to see
each geoflora as a block of tightly defined vegetation was
brought home during a visit I made to Berkeley in 1961. He
had agreed to look at some unidentified material from my
dissertation sites at Trout Creek and Succor Creek in south-
eastern Oregon. These were Miocene floras from the Co-
lumbia Plateau region that, along with numerous associ-
ated floras, were considered typical examples of the
Arcto-Tertiary Geoflora. Chaney had arranged for Harry
MacGinitie to be present to help with the identifications.
[For a biography of MacGinitie, see Wolfe (1987)]. As I
handed each specimen to Chaney, he gave it a cursory
glance and passed it on to MacGinitie, who provided most
of the identifications. The last specimen was a fossil leaf I
suggested might be Arctostaphylos. Chaney did not even
look at this one and passed it to MacGinitie with the com-
ment, "It's not Arctostaphylos" I asked why, and he
replied, "Because Arctostaphylos is not an Arcto-Tertiary
element."
This represents a more rigid view of geofloras than held
by many paleobotanists even at the time. MacGinitie, for ex-
ample, who was a student of Chaney, "questioned whether
any flora, as a unit, migrated during the Tertiary" and wrote
that "The terms 'Arcto-Tertiary,' 'Madro-Tertiary,' and the
like imply extremely useful concepts if we do not think of
these terms as representing areas or centers from which
mass migrations occurred. They picture to us in a general
way the vegetation occupying an area ..." (1982, p. 87).
Studies on the Quaternary history of vegetation have
been particularly instructive on how different combina-
tions of ecologically compatible species can emerge from
successive climatic pertubations (Graham and Grimm,
1990; Webb, 1987). Davis (1976,1981) calculated that dur-
108 Late Cretaceous and Cenozoic History of North American Vegetation
ing the Quaternary, forests seldom maintained a constant
species composition for more than 2000-3000 years and
that a given association may have originated at different
times in different parts of its range. For example, the oak-
chestnut association in the central Applachians "have in-
cluded chestnut as a dominant for 5000 years or longer,
while oak-chestnut forests in Connecticut have included
chestnut for only 2000 years. Deciduous forests in Ohio
were penetrated first by hickory and then by beech 4000
years later; in Connecticut beech arrived first, followed
3000 years later by hickory" (Davis, 1981). In the latest
glacial maximum at 18 Kya, assemblages were present that
have no modern analogs. In the central plains there were
communities variously described as spruce woodland,
spruce forest, prairie, spruce-oak woodland, and black
ash tundra (Davis, 1989). As noted by Wright (1987), these
were grouped into assemblages that are unknown today.
The same applies to several faunal communities where
species whose ranges do not overlap today apparently oc-
curred together during glacial maxima (Graham, 1986; Gra-
ham and Grimm, 1990). The debate regarding the proper
conceptual framework for envisioning Tertiary vegetational
history is paralleled by similar discussions in the early
1900s about its Quaternary history. The Clementsian view
was that plant communities moved mostly as intact units,
while the Gleasonian view was that plants responded to
environmental change as individuals.
If it is accepted that Quaternary climatic changes in-
duced a dynamic response from biotas in terms of their
composition and migratory history, it is instructive to look
at the tempo of these changes as revealed by recent re-
search in glacial geology. The traditional view of northern
hemisphere glacial events was that there were four major
glacial advances (Nebraskan, Kansan, Illinoian, Wiscon-
sin), separated by four interglacials (Aftonian, Yarmouth,
Sangamon, and the present interglacial or Holocene begin-
ning ~11 Kya). Each phase was considered about equal in
duration for the Quaternary (then viewed as encompassing
a little over 1 m.y.), giving a relatively steady pace of -175
Ky for each glacial and interglacial interval. The Quater-
nary has now been extended back to ~ 1.6 Ma, and the pace
of Quaternary climatic change is much quicker than earlier
conceived. There were 18-20 such cycles, with nine oc-
curring within the past 800 Ky (Fig. 3.15; Davis, 1983; John-
son, 1982). If each unit of peak and valley in Fig. 3.15 rep-
resents a time of potential climate-induced change in the
composition of North American plant associations, as was
the case in the latest cycle, then the dynamic nature of
these associations becomes apparent. Similar histories, on
a slower and broader scale, must have extended back into
the pre-Quaternary (Bennett, 1990); it seems unlikely that
any vegetation type has ever migrated as a unit to the ex-
tent implied by the early versions of the geoflora concept.
The question arises as to whether the idea of geofloras
should, or can, be abandoned. Wolfe (1975, 1977, 1994) is
explicit on this point:
"The conclusion is, I think, inescapable that the Arcto-
Tertiary concept has never had a satisfactory strati-
graphic foundation. . . . As knowledge of genetics and
physiology increased, it should have been apparent, as
Mason (1947) pointed out, that the Arcto-Tertiary con-
cept was invalid. The discarding of this concept, which
is indicated by the Alaskan and Siberian assemblages of
fossil plants, is fundamental to an understanding of
floristic and vegetational history." (1977)
Zhilin (1989) concurs, and Boulter (1984) notes that the
theory has confused and delayed proper zonation of the
central European Tertiary. A different view is expressed by
Mai: "For all our palaeophytochronomical studies we con-
firm the old terms 'Arctotertiary' and Talaeotropical' and
also the concept of the geofloras . . ." (1991).
It is likely that geoflora terminology will continue to be
encountered in the literature. The concept is attractive be-
cause it adds a time dimension to extant plant formations.
Also, some designation is convenient for North America
Paleogene floras with prominent tropical elements, Middle
and Late Cenozoic floras characterized more by temperate-
deciduous elements, and Late Cenozoic floras of dry to arid
aspect. Furthermore, just as the geoflora concept eventually
suffered from overdefinition and rigidity, it is also possible,
in marshalling support for its demise, to overstate the ran-
domness of species and generic associations through time.
The results from Quaternary paleoecology previously cited
deal primarily with units of vegetation at the association
level, and it is evident to many paleobotanists that the
broader units of plant formations can be recognized over
long periods of time. For example, Crane et al. acknowl-
edge that the concept oversimplifies the complex history of
plant communities, but add that
"[T]he nearest living relatives of many extinct Paleo-
cene plants are still associated in Recent mixed meso-
phytic forest and this suggests that the climatic and,
perhaps, edaphic tolerances of some individual an-
giosperm lineages have either remained more or less
constant, or have exhibited similar patterns of change
over the last 60 million years." (1990, p. 59)
With regard to the fern Onoclea sensibilis, Rothwell and
Stockey (1991) state that "the species has remained virtu-
ally unchanged in both structural features and ecological
tolerances throughout the Cenozoic" (p. 123) and "thereby
provide one of the most dramatic examples of structural
and ecological stasis known for vascular plants" (p. 114).
Similarly, Webb (1989) refers to intervals of stable faunal
composition that last 10 m.y. or longer and suggests that
many chronofaunas are roughly comparable to extant fau-
nas living in similar situations. Boulter et al. (1993) note
that, "The present-day ecological patterns recognisable in
ecosystems more than twenty million years old suggest
that most of the components [in Miocene plant assem-
blages from Bohemia] retained their ecological preferences
throughout this time." The recent modeling experiments
described in Chapter 2 (CO2 Concentrations section) show

that seasonal variation likely existed in parts of the high
Arctic. This, along with the proposed limitation of tropical
vegetation primarily to a coastal, transported community,
lessens the necessity of viewing temperate and tropical ele-
ments as intermingled and having ecological requirements
completely different from that of their modern descendents.
The geofloras terminology is entrenched in the litera-
ture and likely will continue to be used in the general
sense described by MacGinitie (1962) and by those who
were not involved in or concerned with its tumultuous
past. Moreover, the continued use of a terminology or ap-
plication of a concept that evolves over time is not uncom-
mon. Huggett (1990) presents an interesting discussion of
catastrophism and how its meaning has changed with in-
creasing knowledge of past events in Earth history. The
early proponents of catastrophism like Buckland and Cu-
vier and opponents like Hutton and Lyell would be
amazed at how widely the idea is accepted today and how
little it resembles the original version. Geofloras as a means
of envisioning vegetational history is a heuristic concept;
but if used, it should be in the general sense of MacGinitie
(1962). Other concepts are being forged that emphasize the
importance of Old World tropical lineages in the ancient
high-latitude floras and the contribution of their descen-
dents to the modern vegetation of the New World and that
better reflect the dynamic and temporal nature of plant as-
sociations. One of these is especially relevant to the nature
and history of high-latitude North American vegetation
during the Paleocene.
BOREOTROPICS CONCEPT
Recent studies on Paleogene floras have revealed two char-
acteristics of high-latitude paleocommunities. One is that
even with some limited transport of Cenozoic exotic ter-
ranes along the west coast, there is still evidence for a trop-
ical element associated with temperate deciduous vegeta-
tion in the Arctic Paleogene. The other is that the tropical
component includes plants presently found in Old World
vegetation. The interchange was facilitated by land con-
nections and megathermal climates across the North At-
lantic and Beringia, and the Old World connection is be-
coming more evident with continuing studies of Paleogene
floras. For example, the southeast Asian Sargentodoxa (Sar-
Context 109
Figure 3.15. Oxygen isotope ratios in three deep-sea cores
plotted against time. Low points on the curve represent
times of glacial advance, and high points represent times
of retreat. Arabic numerals designate interglacial stages.
Nine glacial-interglacial cycles are recorded for the past
800,000 years. Reprinted from Davis [1983; based on
Johnson (1982) and Shackleton and Opdyke (1976)] with
the permission of Margaret Davis, the Missouri Botanical
Garden, and the Geological Society of America.
gentodoxaceae) has recently been added to the Early
Miocene Brandon Lignite flora of Vermont (Tiffney, 1993,
1994), and fossil seeds of Ensete (Musaceae; Africa, Asia)
are now known from the Middle Eocene nut bed locality of
the Clarno Formation of north-central Oregon (Manchester
and Kress, 1993). Wolfe (1975) proposed the name boreo-
tropical flora to characterize this mix of high latitude de-
ciduous and evergreen vegetation, unique by the distribu-
tion of extant congeners. It includes some elements that
today are primarily Old World tropical in relationship, as-
sociated with early progenitors of temperate deciduous
vegetation (see also Tiffney, 1985).
In addition to the increasing number of reports of Old
World tropical elements in high-latitude Paleogene floras,
the plausibility of the boreotropical concept can be as-
sessed by testing some of the expected consequences.
Lavin and Luckow (1993) suggest that because South
America was isolated from North America during this
time, "a prediction of this hypothesis posits that a taxon
with a present-day center of diversity in tropical North
America [Mexico, Central America, the Antilles], and with
an early Tertiary fossil record from any region there, has a
high probability of having sister-group relatives in the Pa-
leotropics and derived relatives in South America." They
looked at two taxa of legumes (the Dichrostachys group
and Tribe Robinieae and allies) via area cladograms, and
the results were consistent with the predictions. Further-
more, Wendt (1993) found that ~25% of the tree species in
the Mexican lowland rain forests may be derived from Old
World tropical progenitors that arrived from the north; it
had previously been assumed that virtually all northern
Latin American tropical tree species had come from South
America. The tropical family Bombacaceae may have ar-
rived from the north because the oldest fossil records are
from the lower Maastrichtian of New Jersey (Wolfe, 1975),
and it is not known from South America until the Paleo-
cene. In contrast, Hammel and Zamora (1993) found no ev-
idence for a boreotropical origin of the newly described
Ruptiliocarpon (Lepidobotryaceae) based on the criteria of
Lavin and Luckow (1993). This is apparently one of the
majority of tropical genera that interchanged between
North and South America through the Isthmus of Panama.
More analyses are needed to better quantify the relative
roles of different biogeographic pathways into North
America.
110 Late Cretaceous and Cenozoic History of North American Vegetation

Such data are providing new ways of envisioning the graphic variables in explaining paleoclimates: results
history of North American high-latitude vegetation. With from Cretaceous climate model sensitivity studies. J.
reference to the modern deciduous forest formation, it is Geophys. Res. 89:1267-1279.
now clear that this community is a complex mixture of el- Bennett, K. D. 1990. Milankovitch cycles and their effects
on species in ecological and evolutionary time. Paleo-
ements derived from various sources. Some certainly came biology 16: 11-21.
from mesic Cretaceous ancestors and evolved in the high Berger, W. H. 1982. Climatic steps in ocean history—
northern latitudes. Others are clearly derived from tropical lessons from the Pleistocene. In: Geophysics Study
groups that have evolved temperate components (e.g., Committee, Climate in Earth History. National Acad-
Diospyros virginiana, the persimmon, a New World tem- emy Press, Washington, D.C. pp. 43-54.
perate representative of the primarily tropical family Ebe- Berggren, W. A., D. V. Kent, M.-P. Aubry, and J. Hardenbol
naceae). The contribution of Old World tropical elements (eds.). 1995. Geochronology, Time Scales and Global
Stratigraphic Correlations. SEPM (Society for Sedi-
to the composition of New World temperate deciduous and mentary Geology), Tulsa, OK. Special Publ. 54.
tropical forests is only now becoming fully recognized and Berggren, W. A., D. V Kent, J. D. Obradovich, and C. C.
constitutes support for the concept of a boreotropical flora. Swisher III. 1992. Toward a revised Paleogene chronol-
ogy. In: D. R. Prothero and W. A. Berggren (eds.),
Much of the information from the fossil record is used to re- Eocene-Oligocene Climatic and Biotic Evolution.
construct paleocommunities. For the task of establishing Princeton University Press, Princeton, NJ. pp. 29-45.
the composition, structure, and arrangement of an ancient Berner, R. A. 1994. 3GEOCARB II: a revised model of at-
mospheric CO2 over Phanerozoic time. Am. J. Sci. 294:
plant community, the procedures are complex and beset 56-91.
with potential sources of error and some of the conclusions Berner, R. A., A. C. Lasaga, and R. M. Garrels. 1983. The
and theoretical considerations remain unsettled. If the data carbonate-silicate geochemical cycle and its effect on
are to be used for the even more formidable task of estimat- atmospheric carbon dioxide over the past 100 million
ing paleoenvironments and accurately tracing environmen- years. Am. J. Sci. 283: 641-683.
tal-biotic interactions through time, additional indepen- Birkenmajer, J. 1990. Geochronology and climatostratigra-
dent information is essential. Paleotemperature, sea level, phy of Tertiary glacial and interglacial succession on
King George Island, South Shetland Islands (West
time, and faunal history, along with sound conceptual Antarctica). In: H. Miller (ed.), Zentralblatt fur Geolo-
frameworks, illustrate some of the important contexts nec- gie und Palaontologie Teil I, Workshop on Antarctic
essary, and currently being developed, for accurately inter- Geochronology. E. Schweizerbart'sche Verlagsbuch-
preting the history of North American vegetation. handlung, Stuttgart, pp. 141-151.
Boulter, M. C. 1984. Palaeobotanical evidence for land-
surface temperature in the European Palaeogene. In:
References E J. Brenchley (ed.), Fossils and Climate. Wiley, New
York. pp. 35-47.
Ager, D. V. 1981. Major marine cycles in the Mesozoic. J. , R. Hubbard, and Z. Kvacek. 1993. A comparison of
Geol. Soc. Lond. 138: 159-166. intuitive and objective interpretations of Miocene
Andrews, H. N., Jr. 1980. The Fossil Hunters. Cornell Uni- plant assemblages from north Bohemia. Palaeogeogr.,
versity Press, Ithaca, NY. Palaeoclimatol., Palaeoecol. 101: 81-96.
Axelrod, D. I. 1958. Evolution of the Madro-Tertiary Ge- Brown, J. H. and E. J. Heske. 1990. Control of a desert-
oflora. Botan. Rev. 24: 433-509. grassland transition by a keystone rodent guild. Sci-
. 1981. Altitudes of Tertiary forests estimated from ence 250: 1705-1707.
paleotemperature. Proc. of the Symposium on Qinghai- Chaney, R. W. 1920. The flora of the Eagle Creek Formation.
Xizang (Tibet) Plateau (Beijing, China). Geological and Contrib. Walker Mus. 2: 115-181.
Ecological Studies of Qinghai-Xizang Plateau. Geology, . 1959. Miocene floras of the Columbia Plateau. Part
Geological History and Origin of Qinghai-Xizang I. Composition and Interpretation. Carnegie Inst. Wash.
Plateau. Science Press, Beijing and Gordon & Breach Contrib. Paleontol. 617:1-134.
Science Publishers, New York. Vol. 1, pp. 131-137. -. 1967. Miocene forests of the Pacific Basin; their an-
and H. P. Bailey. 1969. Paleotemperature analysis of cestors and their descendants. Jubilee Publication
Tertiary floras. Palaeogeogr., Palaeoclimatol., Palaeoecol. Commemorating Professor Sasa's 60th Birthday.
6:163-195. Hokkaido University, Sapporo, pp. 209-239.
Barnosky, A. D. 1989. The late Pleistocene event as a para- Chappell, J. and N. J. Shackleton. 1986. Oxygen isotopes
digm for widespread mammal extinction. In: S. K. and sea level. Nature 324:137-140.
Donovan (ed.), Mass Extinctions, Processes and Evi- Coates, A. G. et al. (+ 7 authors). 1992. Closure of the Isth-
dence. Columbia University Press, New York. pp. mus of Panama: the near-shore marine record of Costa
235-254. Rica and western Panama. Geol. Soc. Am. Bull. 104:
Barren, E. J., J. L. Sloan II, and C. G. A. Harrison. 1980. Po- 814-828.
tential significance of land-sea distribution and sur- Crane, P. R., S. R. Manchester, and D. L. Dilcher. 1990. A
face albedo variations as a climatic forcing factor; 180 preliminary survey of fossil leaves and well-preserved
M.Y. to the present. Palaeogeogr., Palaeoclimatol., reproductive structures from the Sentinel Butte Forma-
Palaeoecol. 30:17-40. tion (Paleocene) near Almont, North Dakota. Fieldiana
Barren, E. J. and W. M. Washington. 1984. The role of geo- (Geol. n.s.) 20: 1-63.

Crowley, T. J., J. G. Mengel, and D. A. Short. 1987. Gond-
wanaland's seasonal cycle. Nature 329: 803-807.
Davis, M. B. 1976. Pleistocene biogeography of temperate
deciduous forests. Geosci. Man 13:13-26.
. 1981. Quaternary history and the stability of forest
communities. In: D. C. West, H. H. Shugart, and D. B.
Botkine (eds.), Forest Succession: Concepts and Appli-
cation. Springer-Verlag, Berlin, pp. 132-153.
. 1983. Quaternary history of deciduous forests of
eastern North America and Europe. Ann. Missouri
Bot. Garden. 70: 550-563.
-. 1989. Insights from paleoecology on global change.
Ecol. Soc. Am. Bull. 70: 222-228.
Dawson, M. R., R. M. West, W. Langston, Jr., and J. H.
Hutchison. 1976. Paleogene terrestrial vertebrates:
northernmost occurrence, Ellesmere Island, Canada.
Science 192: 781-782.
Dorf, E. 1964. The use of fossil plants in palaeoclimatic in-
terpretations. In: A. E. M. Nairn (ed.), Problems in
Palaeoclimatology. Interscience Publishers, London,
pp. 13-31.
Eglinton, G. et al. (+ 5 authors). 1992. Molecular record of
secular sea surface temperature changes on 100-year
timescales for glacial terminations I, II, and IV. Nature
356:423-426.
Elias, S. A. 1991. Insects and climatic change. BioScience
41: 552-559.
Emiliani, C. 1955. Pleistocene temperatures. J. Geol. 63:
538-578.
Engler, A. 1879/1882. Versuch einer Entwicklungs-
geschichte der Pflanzenwelt seit der Tertiarperiode
I/II. W. Engelmann, Leipzig, p. 203.
Epstein, S., R. Buchsbaum, H. A. Lowenstam, and H. C.
Urey. 1951. Carbonate-water isotopic temperature
scale. Geol. Soc. Am. Bull. 62: 417-426.
Fairbanks, R. G. 1989. A 17,000-year glacio-eustatic sea
level record: influence of glacial melting rates on the
Younger Dryas event and deep-ocean circulation. Na-
ture 342: 637-642.
FAUNMAP Working Group. 1996. Spatial response of
mammals to Late Quaternary environmental fluctua-
tions. Science 272: 1601-1606.
Geophysics Study Committee. 1990. Sea-Level Change.
National Academy Press, Washington, B.C.
Graham, A. 1965. The Sucker Creek and Trout Creek
Miocene floras of southeastern Oregon. Kent State
Univ. Res. Bull. IX: 1-147.
. 1988. Studies in neotropical paleobotany. VI. The
lower Miocene communities of Panama—the Culebra
Formation. Ann. Missouri Bot. Garden. 75:1467-1479.
-. 1992. Utilization of the isthmian land bridge dur-
ing the Cenozoic—paleobotanical evidence for timing,
and the selective influence of altitudes and climate.
Rev. Palaeobot. Palynol. 72:119-128.
Graham, R. W. 1986. Response of mammalian communities
to environmental changes during the late Quaternary.
In: J. Diamond and T. J. Case (eds.), Community Ecol-
ogy. Harper & Row, New York. pp. 300-313.
and E. C. Grimm. 1990. Effects of global climate
change on the patterns of terrestrial biological commu-
nities. Trends Ecol. Evol. 5: 289-292.
Gray, J. and D. I. Axelrod. 1971. Memorial to Ralph Works
Chaney 1890-1971. Geological Society of America,
Boulder, CO.
Context 111
Guilderson, T. P., R. G. Fairbanks, and J. L. Rubenstone.
1994. Tropical temperature variations since 20,000
years ago: modulating interhemispheric climate
change. Science 263: 663-665.
Gurnis, M. 1990. Ridge spreading, subduction, and sea
level fluctuations. Science 250: 970-972.
. 1993. Phanerozoic marine inundation of conti-
nents driven by dynamic topography above subduct-
ing slabs. Nature 364: 589-593.
Hammel, B. E. and N. A. Zamora. 1993. Ruptiliocarpon
(Lepidobotryaceae): a new arborescent genus and trop-
ical American link to Africa, with a reconsideration of
the family. Novon 3: 408-417.
Haq, B. U, J. Hardenbol, and P R. Vail. 1987. Chronology of
fluctuating sea levels since the Triassic. Science 235:
1156-1166 (See also, Technical Comments. 1988. Sci-
ence, 241: 596-602).
Harrison, C. G. A. 1990. Long-term eustasy and epeirogeny
in continents. In: Geophysics Study Committee (eds.),
Sea-Level Change. National Academy Press, Washing-
ton, D.C. pp. 141-158.
Hays, J. D. and W. C. Pitman III. 1973. Lithospheric plate
motion, sea-level changes and climatic and ecological
consequences. Nature 246:18-22.
Heer, O. 1869. Flora fossilis Alaskana. Kgl. Svenska
Vetensk. Akad. Handl. 8.
Hollick, A. 1936. The Tertiary floras of Alaska. U.S. Geolog-
ical Survey, Denver, CO, Prof. Paper 182, pp. 1-185.
Hsu, K. J. et al. (+ 9 authors). 1977. History of the Mediter-
ranean salinity crisis. Nature 267: 399-403.
Hubbard, R. and M. C. Boulter. 1983. Reconstruction of
Palaeogene climate from palynological evidence. Na-
ture 301: 147-150.
Huggett, R. J. 1990. Catastrophism: Systems of Earth His-
tory. Edward Arnold, London.
Humphrey, S. R. 1974. Zoogeography of the nine-banded
armadillo (Dasypus novemcinctus] in the United
States. BioScience 24: 457-462.
Hutchison, J. H. 1982. Turtle, crocodilian, and champ-
sosaur diversity changes in the Cenozoic of the north-
central region of western United States. Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 37:149-164.
Janis, C. M. 1984. The use of fossil ungulate communties as
indicators of climate and environment. In: R J. Brench-
ley (ed.), Fossils and Climate. Wiley, New York. pp.
85-104.
Johnson, R. G. 1982. Brunhes-Matuyama magnetic rever-
sal dated at 790,000 year B.P by marine-astronomical
correlations. Quatern. Res. 17:135-147.
Kennett, J. R and N. J. Shackleton. 1976. Oxygen isotopic
evidence for the development of the psychrosphere 38
Myr ago. Nature 260: 513-515.
Kerr, R. A. 1984. Carbon dioxide and the control of the ice
ages. Science 223:1053-1054.
Koch, B. E. 1963. Fossil plants from the lower Paleocene of
the Agatdalen (Angmartussut) area, central Nugssauq
Peninsula, northwest Greenland. Medd. Gronland 172.
Koch, P. L., J. C. Zachos, and R D. Gingerich. 1992. Correla-
tion between isotope records in marine and continen-
tal carbon reservoirs near the Palaeocene/Eocene
boundary. Nature 358: 319-322.
Kurten, B. and E. Anderson. 1980. Pleistocene Mammals of
North America. Columbia University Press, New York.
Lavin, M. and M. Luckow. 1993. Origins and relationships
112 Late Cretaceous and Cenozoic History of North American Vegetation
of tropical North America in the context of the
boreotropics hypothesis. Am. J. Bot. 80:1-14.
Leg 119 Shipboard Scientific Party. 1988. Early glaciation
of Antarctica. Nature 333: 303-304.
MacGinitie, H. D. 1962. The Kilgore flora, a late Miocene
flora from northern Nebraska. Univ. Calif. Publ. Geol.
Sci. 35: 67-158.
. 1969. The Eocene Green River flora of northwest-
ern Colorado and northeastern Utah. Univ. Calif. Publ.
Geol. Sci. 83:1-202.
Mai, D. H. 1991. Palaeofloristic changes in Europe and the
confirmation of the Arctotertiary-Palaeotropical ge-
ofloral concept. Rev. Palaeobot. Palynol. 68: 29-36.
Manchester, S. R. and W. J. Kress. 1993. Fossil bananas
(Musaceae): Ensete oregonense sp. nov. from the
Eocene of western North America and its phytogeo-
graphic significance. Amer. J. Bot. 80: 1264-1272.
Martin, R S. and H. E. Wright, Jr. (eds.). 1967. Pleistocene
Extinctions. Yale University Press, New Haven, CT.
Markwick, P. J. 1994. "Equability," continentality, and Ter-
tiary "climate": the crocodilian perspective. Geology
22:613-616.
Mason, H. L. 1947. Evolution of certain floristic associa-
tions in western North America. Ecol. Monogr. 17:
201-210.
Matthews, R. K. and C. Frohlich. 1991. Orbital forcing of
low-frequency glacioeustasy. J. Geophys. Res. 96:
6797-6803.
McKenna, M. C. 1975. Fossil mammals and early Eocene
North Atlantic land continuity. Ann. Missouri Bot.
Garden. 62: 335-353.
. 1980. Eocene paleolatitude, climate, and mammals
of Ellesmere Island. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 30: 349-362.
Miller, K. G., R. G. Fairbanks, and G. S. Mountain. 1987.
Tertiary oxygen isotope synthesis, sea level history,
and continental margin erosion. Paleoceanography 2:
1-19.
Miller, K. G., G. S. Mountain, the Leg 150 Shipboard
Party, and Members of the New Jersey Coastal Plain
Drilling Project. 1996. Drilling and dating New Jer-
sey Oligocene-Miocene sequences: ice-volume,
global sea level, and Exxon records. Science 271:
1092-1095.
Moore, T. C., Jr., T. S. Loutit, and S. M. Greenlee. 1987. Es-
timating short-term changes in eustatic sea level. Pale-
oceanography 2: 625-637.
Moore, T. C., Jr., N. G. Pisias, and L. D. Keigwin, Jr. 1982.
Cenozoic variability of oxygen isotopes in benthic for-
maminifera. In: Geophysics Study Committee (eds.),
Climate in Earth History. National Academy Press,
Washington, D.C. pp. 172-182.
Opdyke, N. D. 1990. Magnetic stratigraphy of Cenozoic ter-
restrial sediments and mammalian dispersal. J. Geol.
98: 621-637.
Pitman W. C., III. 1978. Relationship between eustacy and
stratigraphic sequences of passive margins. Geol. Soc.
Am. Bull. 89: 1389-1403.
Poore, R. Z. and R. K. Matthews. 1984. Oxygen isotope
ranking of Late Eocene and Oligocene planktonic
foraminifera: Implications for Oligocene sea-surface
temperatures and global ice-volume. Mar. Micropale-
ontol. 9:111-134.
Prentice, K. C. and I. Y. Fung. 1990. The sensitivity of ter-
restrial carbon storage to climate change. Nature 346:
48-51.
Prentice, M. L. and R. K. Matthews. 1988. Cenozoic ice-
volume history: development of a composite oxygen
isotope record. Geology 16: 963-966.
. 1991. Tertiary ice sheet dynamics: the snow-gun
hypothesis. J. Geophys. Res. 96: 6811-6827.
Prothero, D. R. 1994. The late Eocene-Oligocene extinc-
tions. Annu. Rev. Earth Planet. Sci. 22: 145-165.
and C. C. Swisher III. 1992. Magnetostratigraphy
and geochronology of the terrestrial Eocene-
Oligocene transition in North America. In: D. R.
Prothero and W. A. Berggren (eds.), Eocene-Oligocene
Climatic and Biotic Evolution. Princeton University
Press, Princeton, NJ. pp. 46-73.
Roemmich, D. 1992. Ocean warming and sea level rise
along the southwest U.S. Coast. Science 257: 373-375.
Rothwell, G. W. and R. A. Stockey. 1991. Onoclea sensibilis
in the Paleocene of North America, a dramatic exam-
ple of structural and ecological stasis. Rev. Palaeobot.
Palynol. 70:113-124.
Savage, D. E. and D. E. Russell. 1983. Mammalian Paleofau-
nas of the World. Addison-Wesley, London.
Savin, S. M. 1977. The history of the Earth's surface tem-
perature during the past 100 million years. Annu. Rev.
Earth Planet. Sci. 5: 319-355.
. 1982. Stable isotopes in climatic reconstructions.
In: Geophysics Study Committee (eds.), Climate in
Earth History. National Academy Press, Washington,
D.C. pp. 164-171.
and R. G. Douglas. 1985. Sea level, climate, and the
Central American land bridge. In: E G. Stehli and S. D.
Webb (eds.), The Great American Biotic Interchange.
Plenum, New York. pp. 303-324.
Savin, S. M. and E Woodruff. 1990. Isotopic evidence for
temperature and productivity in the Tertiary oceans.
In: W. C. Burnett and S. R. Riggs (eds.), Phosphate De-
posits of the World, Neogene to Modern Phosphorites.
Cambridge University Press, Cambridge, U.K. Vol. 3,
pp. 241-259.
Schlanger, S. O., H. C. Jenkyns, and I. Premoli-Silva. 1981.
Volcanism and vertical tectonics in the Pacific Basin
related to global Cretaceous transgressions. Earth
Planet. Sci. Lett. 52: 435-449.
Schoell, M., S. Schouten, J. S. Sinninghe Damste, J. W. de
Leeuw, and R. E. Summons. 1994. A molecular organic
carbon isotope record of Miocene climate changes.
Science 263: 1122-1125.
Schrag, D. P, G. Hampt, and D. W. Murray. 1996. Pore fluid
constraints on the temperature and oxygen isotopic
composition of the glacial ocean. Science 272:
1930-1932.
Shackleton, N. J. 1974. Attainment of isotopic equilibrium
between ocean water and the benthonic foraminiferal
genus Uvigerina: isotopic changes in the ocean during
the last glacial. Colloq. Int. C.N.R.S. 219: 203-210.
. 1987. Oxygen isotopes, ice volume and sea level.
Quatern. Sci. Rev. 6:183-190.
and A. Boersma. 1981. The climate of the Eocene
Ocean. J. Geol. Soc. Lond. 138: 153-157.
Shackleton, N. J. and J.-C. Duplessy. 1986. Temperature
changes in ocean deep waters during the late Pleis-
tocene. 2nd International Conference on Paleoceanog-
raphy, Woods Hole, MA. Abstract.

Shackleton, N. J. and J. E Kennett. 1975a. Paleotemperature
history of the Cenozoic and the initiation of Antarctic
glaciation: oxygen and carbon isotope analyses in
DSDP sites 277, 279 and 281. Init. Rep. Deep Sea Drill.
Proj., 29: 743-755.
. 1975b. Late Cenozoic oxygen and carbon isotopic
changes at DSDP site 284: implications for glacial his-
tory of the northern hemisphere and Antarctica. Initial
Rep. Deep Sea Drilling Project, 29: 801-807.
Shackleton, N. J. and N. D. Opdyke. 1976. Oxygen-isotope
and paleomagnetic stratigraphy of equatorial Pacific
core V28-239 Late Pliocene to latest Pleistocene. In:
R. M. Cline and J. D. Hays (eds.), Investigation of Late
Quaternary Paleoceanography and Paleoclimatology.
Geological Society of America, Boulder, CO. Mem.
145, pp. 449-464.
. 1977. Oxygen isotope and palaeomagnetic evi-
dence for early northern hemisphere glaciation. Na-
ture 270: 216-219.
Sheehan, P. M., D. E. Fastovsky, R. G. Hoffmann, C. B.
Berghaus, and D. L. Gabriel. 1991. Sudden extinction
of the dinosaurs: latest Cretaceous, upper Great Plains,
U.S.A. Science 254: 835-839.
Shotwell, J. A. 1968. Miocene mammals of southeast Ore-
gon. Univ. Oregon Mus. Nat. Hist. Bull. 14.
Southam, J. R. and W. W. Hay. 1977. Time scales and dy-
namic models of deep-sea sedimentation. J. Geophys.
Res. 82: 3825-3842.
. 1981. Global sedimentary mass balance and sea
level changes. In: C. Emiliani (ed.), The Ocean Litho-
sphere. Wiley, New York. pp. 1617-1684.
Stehli, E G. and S. D. Webb (eds.). 1985. The Great Ameri-
can Biotic Interchange. Plenum, New York.
Taggart, R. E. and A. T. Cross. 1990. Plant successions
and interpretations in Miocene volcanic deposits,
Pacific northwest. In: M. G. Lockley and A. Rice
(eds.), Volcanism and Fossil Biotas. Geological Soci-
ety of America, Boulder, CO. Special Paper 244, pp.
57-68.
, A. T. Cross, and L. Satchell. 1982. Effects of peri-
odic volcanism on Miocene vegetation distribution in
eastern Oregon and western Idaho. Proc. 3rd North
Am. Paleontol. Conv. 2: 535-540.
Tanai, T. and K. Huzioka. 1967. Climatic implications of
Tertiary floras in Japan. Symposium 25, llth Pacific
Science Congress, University of Tokyo, August-
September, 1966, Tertiary Correlations and Climatic
Changes in the Pacific. Tokyo, Japan, pp. 89-94.
Thompson, S. L. and E. J. Barron. 1981. Comparison of Cre-
taceous and present Earth albedos: implications for
the causes of paleoclimates. J. Geol. 89: 143-167.
Tiffney, B. H. 1985. The Eocene North Atlantic land bridge:
its importance in Tertiary and modern phytogeography
of the northern hemisphere. J. Arnold Arb. 66:
243-273.
. 1993. Fruits and seeds of the Tertiary Brandon lig-
nite. VII. Sargentodoxa (Sargentodoxaceae). Am. J.
Bot. 80: 517-523.
-. 1994. Re-evaluation of the age of the Brandon lig-
nite (Vermont, USA) based on plant megafossils. Rev.
Palaeobot. Palynol. 82: 299-315.
United States Department of Agriculture Yearbook. 1941.
Climate and Man. U.S. Gov. Printing Office, Washing-
ton, D.C.
Context 113
Urey, H. C. 1947. The thermodynamic properties of iso-
topic substances. J. Chem. Soc. Lond., 1947: 562-581.
, H. A. Lowenstam, S. Epstein, and C. R. McKinney.
1951. Measurement of paleotemperatures and temper-
atures of the Upper Cretaceous of England, Denmark,
and the southeastern United States. Geol. Soc. Am.
Bull. 62: 399-416.
Vail, R R. and J. Hardenbol. 1979. Sea-level changes during
the Tertiary. Oceanus 22: 71-79.
Vail, R R., R. M. Mitchum, Jr., and S. Thompson III. 1977.
Seismic stratigraphy and global changes of sea level,
part 4: global cycles of relative changes of sea level. In:
C. E. Payton (ed.), Seismic Stratigraphy—Applications
to Hydrocarbon Exploration. American Association of
Petroleum Geologists, Tulsa, OK. Mem. 26, pp. 83-97.
Warrick, R. A., E. M. Barrow, and T. M. L. Wigley. 1993. Cli-
mate and Sea Level Change: Observations, Projections,
and Implications. Cambridge University Press, Cam-
bridge, U.K.
Watts, A. B. 1982. Tectonic subsidence, flexure and global
changes of sea level. Nature 297: 469-474.
Webb, S. D. 1969. Extinction-origination equilibria in Late
Cenozoic land mammals of North America. Evolution
23:688-702.
. 1977. A history of savanna vertebrates in the New
World. Part I: North America. Annu. Rev. Ecol. Syst. 8:
355-380.
. 1985. Main pathways of mammalian diversifica-
tion in North America. In: F. G. Stehli and S. D. Webb
(eds.), The Great American Biotic Interchange.
Plenum, New York. pp. 201-217.
. 1989. The fourth dimension in North American ter-
restrial mammal communities. In: D. W. Morris, Z.
Abramsky, B. J. Fox, and M. R. Willig (eds.), Patterns in
the Structure of Mammalian Communities. Museum of
Texas Tech University, Lubbock, TX. Special Publ. 28,
pp. 181-203.
-, R. C. Hulbert, Jr., and W. D. Lambert. 1995. Climatic
implications of large-herbivore distributions in the
Miocene of North America. In: E. S. Vrba, G. H. Den-
ton, T. C. Partridge, and L. H. Burckle (eds.). 1995. Pa-
leoclimate and Evolution, with Emphasis on Human
Origins. Yale University Press, New Haven, CT. pp.
91-108.
Webb T., III. 1987. The appearance and disappearance of
major vegetational assemblages: long-term vegeta-
tional dynamics in eastern North America. Vegetatio
69: 177-187.
Wendt, T. 1993. Composition, floristic affinities, and ori-
gins of the canopy tree flora of the Mexican Atlantic
slope rain forests. In: T. P. Ramamoorthy, R. Bye, A.
Lot, and J. Fa (eds.), Biological Diversity of Mexico:
Origins and Distribution. Oxford University Press, Ox-
ford, U.K. pp. 595-680.
West, R. M. and M. R. Dawson. 1978. Vertebrate paleontol-
ogy and the Cenozoic history of the North Atlantic re-
gion. Polarforschung 48: 1103-1109.
Westgate, J. W. and C. T. Gee. 1990. Paleoecology of a mid-
dle Eocene mangrove biota (vertebrates, plants, and in-
vertebrates) from southwest Texas. Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 78:163-177.
Wing, S. L., T. M. Bown, and J. D. Obradovich. 1991. Early
Eocene biotic and climatic change in interior western
North America. Geology 19: 1189-1192.
114 Late Cretaceous and Cenozoic History of North American Vegetation
Wolfe, J. A. 1971. Tertiary climatic fluctuations and meth-
ods of analysis of Tertiary floras. Palaeogeogr., Palaeo-
climatol., Palaeoecol. 9: 27-57.
. 1975. Some aspects of plant geography of the
northern hemisphere during the Late Cretaceous and
Tertiary. Ann. Missouri Bot. Garden. 62: 264-279.
. 1977. Paleogene floras from the Gulf of Alaska re-
gion. U.S. Geological Survey, Denver, CO, Professional
Paper 997, pp. 1-108.
. 1978. A paleobotanical interpretation of Tertiary
climates in the northern hemisphere. Am. Sci. 66:
694-703.
. 1987. Memorial to Harry D. MacGinitie (1896-
1987). Ann. Missouri Bot. Garden. 74: 684-688.
. 1994. Alaskan Paleogene climates as inferred from
the CLAMP database. In: M. C. Boulter and H. C.
Fisher (eds.), Cenozoic Plants and Climates of the Arc-
tic. NATO ASI Series 1: Global Environmental Change.
Springer-Verlag, Berlin. Vol. 27, pp. 223-238.
and D. M. Hopkins. 1967. Climatic changes
recorded by Tertiary land floras in northwestern North
America. In: K. Hatai (ed.), Tertiary Correlations and
Climatic Changes in the Pacific. Sendai, Sasaki, Japan,
pp. 67-76.
Wolfe, J. A., and R. Z. Poore. 1982. Tertiary marine and
nonmarine climatic trends. In: Geophysics Study
Committee (eds.), Climate in Earth History. National
Academy Press, Washington, D.C. pp. 154-158.
Woodburne, M. O. (ed.). 1987. Cenozoic Mammals of North
America: Geochronology and Biostratigraphy. Univer-
sity of California Press, Berkeley, CA.
Wright, H. E., Jr. 1987. Synthesis; the land south of the ice
sheets. In: W. E Ruddiman and H. E. Wright, Jr. (eds.),
North America and Adjacent Oceans during the Last
Deglaciation. DNAG (Decade of North American Geol-
ogy) Geological Society of America, Boulder, CO. Vol.
K-3, pp. 479-488.
Ye, Q., R. K. Matthews, W. E. Galloway, C. Frohlich, and S.
Gan. 1993. High-frequency glacioeustatic cyclicity in
the early Miocene and its influence on coastal and
shelf depositional systems, NW Gulf of Mexico Basin.
In: J. M. Armentrout, R. Bloch, H. C. Olson, and B. E
Perkins (eds.), Rates of Geologic Processes: Tectonics,
Sedimentation, Eustasy and Climate: Implications for
Hydrocarbon Exploration. 14th Annual Research Con-
ference, Gulf Coast States Section, Society of Eco-
nomic Paleontology and Mineralogy Foundation, Tulsa,
OK. pp. 287-298.
Yemane, K. 1993. Contribution of Late Permian palaeo-
geography in maintaining a temperate climate in
Gondwana. Nature 361: 51-54.
Zhilin, S. G. 1989. History of the development of the tem-
perate forest flora in Kazakhstan, U.S.S.R. from the
Oligocene to the early Miocene. Bot. Rev. 55: 205-330.
General Readings
Anderson, D. M. and R. S. Webb. 1994. Ice-age tropics re-
visited. Nature 367: 23-24.
Broecker, W. S. 1996. Glacial climate in the tropics. Science
272: 1902-1904.
Kerr, R. A. 1980. Changing global sea levels as a geologic
index. Science 209: 483-486.
. 1984. Vail's sea-level curves aren't going away. Sci-
ence 226: 677-678.
. 1987. Refining and defending the Vail sea level
curve. Science 235:1141-1142.
. 1994. In tropical coral, signs of an ice age chill. Sci-
ence 263: 173.
. 1996. Ancient sea-level swings confirmed. Science
272: 1097-1098.
Lyle, M. 1992. Molecules record sea change. Nature 356:
385-386.
Warrick, R. A. 1993. Sea-level rise. EarthQuest, Office for
Interdisciplinary Earth Studies, University Corpora-
tion for Atmospheric Research, Boulder, CO.
Notes
1. The standard selected was the ratio in a fossil belem-
nite from the Cretaceous Pee Dee Formation of North Car-
olina. The standard had a high isotopic Content and val-
ues are expressed as percent negative compared to the
standard.
2. A linear regression is the conventional mathematical
method for data derivation from values for constants used
in expressions of simple (linear) relationships that involve
two variables (in this case, fluctuating temperatures and
changing isotope concentrations).
 FOUR

Methods, Principles, Strengths,

and Limitations

Methods of paleovegetation analysis can be grouped into began to routinely apply paleopalynology to problems of
two broad categories. Those that use plant microfossils for stratigraphic correlation and the reconstruction of deposi-
reconstructing terrestrial vegetation, past environments, tional environments in Tertiary and older strata (Hoffmeis-
migrations, and evolutionary histories constitute a part of ter, 1959). This added a practical dimension to a mostly
paleopalynology that includes the study of pollen, spores, academic pursuit and fostered interest in applied palynol-
other acid-resistant microscopic structures, and phytoliths ogy and its use as a paleoecological research tool. This im-
(distinctive, microscopic silicate particles produced by portant development is reflected in the increased number
vascular plants). Those that use plant megafossils such as of publications after about 1955 (Fig. 4.1; Jansonius and
leaves, cuticles, cones, flowers, fruits, and seeds constituteMcGregor, 1996). As the history of other innovations might
paleobotany. Two important subdisciplines of paleobotany predict, there was a period of exuberant claims, isolated
are dendrochronology (fossil woods) and analysis of pack- specialization, and exaggerated charges of deficiency in the
rat middens. The latter are sequences of nesting materials, method; but for palynology this seemingly inevitable pe-
constructed by packrats of the genus Neotoma, preserved riod was mercifully brief. The different terminology, prin-
in arid environments of the American southwest. The ciples, and techniques involved in megafossil paleobotany
study of fossil fruits and seeds is a specialized field withinand paleopalynology still result in specialization, but this
paleobotany, and it is also used in studies on Quaternary limitation is frequently overcome by coordinated or collab-
vegetational history in the preparation of seed diagrams ac- orative projects, and an increasing number of practitioners
companying pollen and spore profiles from bog and lake work in both disciplines. Palynology is now one of several
sequences. important sources of data available for decifering the com-
plex interactions between environmental change and veg-
etational history.
PALEOPALYNOLOGY (PLANT MICROFOSSILS) The technique is based on several principles or general-
izations. One is that essentially all plants produce either
pollen grains or spores. This means that with few excep-
Pollen and Spores
tions (e.g., Lauraceae, some aquatics), all taxonomic or eco-
In 1916 Swedish geologist Lennart von Post demonstrated logic groups of plants are potentially represented in the
that pollen grains and spores were abundantly preserved fossil record.
in Quaternary peat deposits and could be used to trace re- The second basis for the technique is that these pollen
cent forest history and climatic change (Davis and Faegri, grains and spores are released to the atmosphere and fall as
1967). The term palynology was subsequently introduced the pollen and spore rain. During transport, mixing occurs
by Hyde and Williams in 1944 to include all studies con- in the atmosphere and the settling particles provide a re-
cerned with pollen and spores. Paleopalynology has come gional picture of the surrounding vegetation. In the early
to denote the study of acid-resistant microfossils generally, days of palynology it was voguish to point out that pollen
while pollen analysis designates those investigations deal- and spores could be "blown in from anywhere" and that
ing specifically with the Quaternary. microfossil assemblages were an ecologically meaningless
In the early 1950s researchers in the petroleum industry mixture of elements from various communities and distant

115
116 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 4.1. Number of papers on
pre-Pleistocene palynology published
between 1940 and 1966. Note the in-
crease in the early 1950s correlated with
the beginnings of extensive use of
palynology in the petroleum industry.
Reprinted from Kremp and Methvin
(1968) with the permission of the
Oklahoma Geological Survey.
regions. This is not the case; in fact, most pollen and spores
fall within a few miles or less of the parent plant (see re-
view in Farley, 1990). The exceptions, such as pine, are
well known to palynologists (Kapp, 1961; Ritchie and
Lichti-Federovich, 1967) and long-distance transport can
be compensated for in pollen and spore assemblages.
The local to moderate aerial distribution of most pollen
grains and spores can reveal the principal elements of up-
land associations growing some distance from the deposi-
tional basin and elements of understory vegetation. These
plants are commonly absent or less well represented in the
megafossil record. The pollen of several temperate genera
has been recovered from Eocene floras in the Mississippi
Embayment region, which on the basis of the megafossils,
preserving mostly strand plants, was previously inter-
preted as coastal lowland tropical rain forest. The inland
temperate component recorded by the pollen flora reveals
a more complex landscape, a greater variety of habitats,
and a warm temperate to subtropical climate.
Another positive result of the aerial distribution of
pollen and spores is their incorporation into a variety of ac-
cumulating sediments. Like a synchronous ash fall, these
microscopic particles settle into near-shore marine habitats
and onto lake, bog, swamp, and floodplain surfaces. Thus,
they may be recovered from sediments representing ma-
rine, brackish water, freshwater, and a variety of terrestrial
habitats. Atmospheric mixing, rather than a deficiency, is
one of the strengths of paleopalynology in reconstructing
paleocommunities and past environments.
The third principle of paleopalynology is that pollen
grains and spores are capable of being fossilized. Their
durability is a result of the chemical composition of the
wall. The chemistry is still not perfectly understood, but in
pollen the exine (pollen wall) is probably composed of ox-
idative polymers of carotenoids and/or carotenoid esters
called sporopollenin; the approximate empirical formula
of which is CgoH142O27 (Brooks et al., 1971; Shaw, 1971).
This substance is resistant to degradation under acid con-
ditions but is destroyed by bases, extreme heat and pres-
sure, mechanical corrosion, crystallization, enzymatic dis-
solution by fungi and bacteria, and by oxidation. As a
result, well-preserved pollen and spores are not normally
found in alkaline, igneous, metamorphic, or coarse-grained
rocks because of sorting and oxidation; in sediments that
accumulated very slowly or were agitated (aerated) for long
periods before settling; or in deposits that accumulated
under arid conditions or were exposed to postdepositional
drying. The most favorable depositional environments are
acidic, stagnant, poorly oxygenated, and/or with a rapid
accumulation of fine-grained particles. Sediments de-
posited under these conditions are lithified into near-shore
marine and deltaic shales; lacustrine volcanic shales, mud-
stones, claystones, and siltstones; and terrestrial (swamp)
peat, lignite, and coal.
The variety of sediment types in which pollen and
spores occur is the principal basis for their value in strati-
graphic correlation. Most plants and animals are restricted
to one kind of habitat (e.g., marine, freshwater, or terrestrial)
and their fossils are found only in one or a few correspond-
ing types of sediment. This makes correlation difficult be-

tween rocks of different lithologies on the basis of fossil
content. The problem in geologic parlance is called trans-
gressing the marine-nonmarine boundary. Pollen grains
and spores can be used to correlate a wide variety of strata
representing many kinds of depositional environments.
Because of their durability, acid-resistant plant micro-
fossils are preserved in strata ranging in age from Precam-
brian to the present, and no measurable segment of time
after -3.8 billion years is unrepresented in the plant micro-
fossil record. Also, palynomorphs can often be recovered
from strata that lack plant megafossils, thus expanding the
potential data base for studies on vegetational history.
The fourth principle is that techniques are available for
the recovery of pollen and spores from the sediments.
These techniques involve processing the samples through
a series of acids (HC1, HF, HNO3) that dissolve or oxidize
various mineral (carbonate, silicate) and nonsporopollenin
components (e.g., lignins), leaving a concentrated residue
of acid-resistant microfossils (Gray, 1965; Traverse, 1988).a
The fifth principle or generalization is that pollen and
spores vary in morphology such that plants can be recog-
nized on the basis of their pollen and spore characters. The
taxonomic level to which they can be recognized differs
among various plant groups, but it is frequently to genus.
There are some notable exceptions. In a few genera, such as
Cuphea (Fig. 4.2) in the Lythraceae and Bauhinia in the
caesalpinioid legumes (Fig. 4.3), many pollen types can be
recognized to the species level. In many more instances,
however, identifications can be made only to the family
level [Cyperaceae (sedges), Poaceae-Gramineae (grasses)]
or to groups of families (Chenopodiaceae, Amaranthaceae;
Blechnaceae, Polypodiaceae, Pteridaceae). In families such
as the Compositae-Asteraceae, Leguminosae-Fabaceae,
Orchidaceae, Palmae-Areaceae, and others there is con-
siderable pollen variation, but this is countered by the
large size of the families, and only in certain instances can
generic identifications be made. The occurrence of plant
microfossils at localities that also preserve megafossils pro-
vides leads for the identification of unknowns in both
groups.
Many fossil pollen grains and spores resemble those of
modern genera back through about the Middle Eocene. In
older deposits an increasing number of specimens are remi-
niscent of certain extant genera and families, but they can-
not be assigned with certainty to modern taxa. The generic
affinities of some earlier gymnosperm pollen and fern spores
can be recognized, but with increasing age the likelihood
that they are comparable biologically and ecologically to ex-
tant genera becomes more remote. The modern analog
method for reconstructing paleoenvironments, based on fos-
sil pollen and spore assemblages, is applicable to deposits
from the present back through about the Middle Eocene.
The sixth principle is that plants are limited in their eco-
logical tolerance. An important corollary to this fact is that
therefore paleoecological conditions can be reconstructed
on the basis of fossil pollen and spore assemblages.
Methods, Principles, Strengths, and Limitations 117
An area where paleopalynology has proven especially
useful is in providing a record of evergreen, herbaceous,
annual, and suffrutescent plants (those that die back each
year to a perrenial root stock). In most instances the leaves
of these plants wither on the stem, or they are not shed in
sufficient quantities or widely enough distributed to be ex-
tensively represented in the megafossil record. Fruits,
seeds, and flowers are occasionally recovered and are of
great value in biogeography and for tracing the evolution-
ary history of individual lineages. However, there are few
such assemblages and the fossils are not usually found in
sufficient quantities to make paleocommunity and envi-
ronmental reconstructions. In contrast, many of these
plants produce large amounts of pollen and spores, which
provide a record of herbaceous and understory vegetation
that otherwise would be meager to nonexistant. Such a
record is important for providing an accurate inventory of
plants from fossil floras and a true picture of plant diver-
sity for a region, detecting the disturbance of vegetation as-
sociated with the early introduction of agriculture, tracing
the time and direction of movement of ancient cultures
into a new region, and determining the time and place of
origin of important cultivars. The presence of pollen in the
anthers of ancient flowers can also aid in evaluating taxo-
nomic affinities (e.g., Crepet et al., 1992).
These principles impart to paleopalynology its particu-
lar set of strengths, as well as certain limitations. There are
no morphological features of pollen or spores consistently
associated with particular habitats or climate and therefore
no "pollen or spore physiognomy" to aid in paleoenviron-
mental reconstructions. As a result, paleopalynological
studies used for tracing vegetational history are limited to
the modern analog method. As noted previously, pollen
and spores can often be identified to the genus level, rarely
to species, and in many instances only to a group of genera
or higher taxonomic units. Thus, within the modern analog
context, paleocommunity and paleoenvironmental recon-
structions are further limited to inventories consisting
mostly of generic identifications.
There are other aspects of the paleopalynological method
that limit its precision in estimating the qualitative and
quantitative characteristics of paleocommunities. Certain
pollen and spore types lack sporopollenin or have thin ex-
ines that are destroyed in the fossilization and recovery
processes, and they are frequently absent or underrepre-
sented in fossil assemblages. Examples include the rela-
tively delicate pollen ofPopulus, some Juncaceae (rushes),
and Cyperaceae. Even when present, these grains are often
folded and crumpled to the extent that identification and
accurate counts are difficult. In some plant groups the wall
is so delicate or the chemical composition is such that the
grains are seldom preserved. The pollen of virtually all
members of the order Zingiberales, with the exception of a
few genera in the Costaceae (e.g., Tapeinochilos; Stone et
al., 1979, 1981), is destroyed during fossilization or in the
processing procedure. Of special significance in this regard
 m
Figure 4.2. Light photomicrographs illustrating diversity in pollen morphology at the species level in Cuphea (family
Lythraceae). (A) C patula (25 urn), (B) C. urbaniana (28 urn), (C) C. purpurescens (28 urn), (D) C. retroscabra (32 urn)
(E) C. koehneana (28 urn), (F) C. reitzii (32 urn), (G) C. parsonsia (25 x 32 urn), (H) C. pseudosilene (38 urn), (I) Cfer-
nsiae (32 urn), (J) C. bustamanta (36 Um), (K) C. campestris (32 urn), (L) C. viscosa (42 pm), (M) C. glossostoma (4Q um)
 \\o.

Figure 4.3. Scanning electron microscope photomicrographs illustrating diversity in pollen morphology at the
species level in Bauhinia (family Leguminosae). (A,B) B. aculata (original magnifications x851, x2500), (C,D)
B. divaricata (original magnifications x!750, xSOOO), (E,F) B. dipetala (original magnifications xl!40, x3170).
119
120 Late Cretaceous and Cenozoic History of North American Vegetation
is pollen of the family Lauraceae. It is a relatively minor
component of the present temperate flora [Lindera, Litsea,
Nectandra, Sassafras, Persea, Umbellaria (California lau-
rel)], but the family is prominent in extant tropical vegeta-
tion and in the Early to Middle Cenozoic floras of North
America. During the Late Cretaceous and Early Tertiary,
when tropical communities extended into north temperate
regions, or when exotic terranes were transported northward
(e.g., the Miocene Carmel flora of California; D. I. Axelrod,
personal communication, 1996), members of the Lauraceae
were present, as documented by the megafossil record (Dil-
cher, 1963; Herendeen, 1991), but they are not found in
plant microfossil assemblages. In contrast, thick-walled
spores and pollen are often overrepresented in fossil assem-
blages. Many fern spores and the heavy-walled pollen of the
Malpighiaceae and other groups are occasionally found to
the near exclusion of thin-walled types. These examples col-
lectively constitute differential preservation and this is one
of the limitations of the paleopalynological method.
Another facet of the technique that limits quantitative
reconstructions is differential production. Plants vary
widely in the amount of pollen and spores produced,
which complicates the estimation of the relative abun-
dance of various components of the paleocommunity from
microfossil assemblages. Differential production can be
compensated for to some extent in Quaternary deposits by
knowing the approximate amounts of pollen produced by
modern plants. Using figures cited in Faegri et al. (1989),
some estimates are
Phoenix dactylifera (date palm): 89,000 grains/anther
Cannabis (hemp): 70,000 grains/anther
Betula: 10,000 grains/anther
Acer: 1000 grains/anther
Linum catharticum (flax): 100 grains/anther
Malva: 64 grains/anther
10-year-old branch ofPinus sylvestris (scotch pine):
350 million grains
10-year-old branch ofPicea, Betula, Quercus:
100 million
10-year-old branch ofFagus sylvatica: 28 million
Spruce forests of southern Sweden: 75,000 tons/year
Pinus sylvestris: 10-80 kg/ha/year (=-30,000-280,000
grains/cm2/season)
Rybinsk Reservoir (USSR) vegetation: 6 kg/ha/year
Another factor in assessing the relative abundance of el-
ements in a paleocommunity from pollen and spore evi-
dence is the spatial arrangement of the species. A few indi-
viduals growing marginal to the basin of deposition may
leave a numerical record similar to that from a larger num-
ber of individuals growing farther away. In Pleistocene and
Holocene deposits accurate estimates of relative abun-
dance and spatial arrangement can be obtained from mod-
ern pollen rain studies from different vegetation types, a
knowledge of the ecology and composition of comparable
modern associations, and by numerical methods of analy-
sis such as pollen influx and absolute pollen frequencies
(Chapter 8). For older sediments only general approxima-
tions of the composition, abundance, and arrangement of
associations within plant formations can be gained from
plant microfossil assemblages. The pollen of anemophi-
lous (wind-pollinated) species is numerically overrepre-
sented in pollen assemblages, while entomophilous (insect-
pollinated) species are underrepresented. For example, all
of the common pollen types recovered in one study of the
latest Eocene to earliest Oligocene Florissant flora of Col-
orado were wind pollinated (Fig. 4.4).
In addition to limitations, there are other aspects of the
method that constitute potential sources of error. These
have been addressed extensively in the literature and are
summarized by Faegri et al. (1989) and Traverse (1988).
Contamination is the most important and it can occur dur-
ing and after deposition and during the collection and pro-
cessing of samples.
Redeposition is the process whereby sediments are
eroded by wind or water, carried into a new depositional
basin (allochthonous material), and deposited along with
newly accumulating sediments (autochthonous material;
Frederiksen, 1989). The exotic material usually represents
a different age and often a different kind of environment.
Some depositional conditions are particularly susceptible
to contamination through redeposition. Deltas receive sed-
iments transported by rivers that often erode a variety of
substrates, and the resulting shales can contain a mixture
of palynomorphs of widely different ages.
There are several ways to detect redeposited specimens.
One is by noting the quality of preservation. Specimens
that have been redeposited are frequently corroded or frag-
mented. The color of the microfossils usually darkens with
age, and the presence of noticeably darker or lighter grains
than is typical for the overall assemblage suggests possible
contamination. The petroleum industry uses color charts
to estimate the thermal maturity of organic matter (e.g.,
Pearson, 1984). At a certain critical temperature, referred to
as the "window of petroleum formation," droplets of oil
begin to form from kerogen. The older and most thermally
altered specimens are darker, while younger palynomorphs
are progressively lighter in color. The color charts, de-
signed to estimate thermal maturity, also give a rough esti-
mate of the relative age of different components in the sam-
ple. Another clue to redeposited pollen and spores is the
morphology of the specimens. If palynomorphs of pre-
Middle Eocene age have been redeposited into younger
strata, experienced palynologists familiar with the section
can often detect the contaminants through a subtle combi-
nation of morphological features atypical of younger spec-
imens. These three characteristics of preservation quality,
color, and morphology provide clues to possible contami-
nation. A knowlege of the depositional environment for
different rock types, the stratigraphic range of various paly-
nomorphs, and a familiarity with the fossil content of adja-
cent strata can also aid in assessing the likelihood of rede-
position. Cracks, fissures, and large sinks allow sediments

Figure 4.4. Chart of common pollen and spore types from the Late Eocene Florissant Beds National Monument, Colorado.
All common pollen taxa are from wind-pollinated woody plants. Reprinted from Leopold et al. (1992) with the permis-
sion of Princeton University Press.
of one age to infiltrate older strata. The presence of these
structures is usually revealed by abrupt horizontal changes
in lithology.
A more subtle source of potential contamination, until
it was brought to the attention of palynologists, was from
drilling mud thinners (Traverse et al., 1961). At one of the
early meetings of the American Association of Strati-
graphic Palynologists I was asked to present a summary of
work on the Tertiary history of vegetation in northern Latin
America. One facet of this presentation dealt with the pro-
gressively later appearance of northern temperate pollen
types into Mexico, Central America, and northern South
America (Alnus, Ilex, Myrica, Quercus, and others). At the
end of the presentation a palynologist working for one of
the oil companies informed the audience that the pur-
ported trends were figments of the imagination and that
these temperate types were widespread during the entire
Cenozoic throughout the tropics of the western hemi-
sphere. That has proved to not be the case. One possible
explanation for the confusion was down-hole contamina-
tion of cuttings and side-wall contamination of cores from
drilling mud thinners. This source of error was not univer-
sally recognized in the early days of palynology. During
drilling operations a mud or flux is circulated into the drill
Methods, Principles, Strengths, and Limitations 121
hole as a lubricant to prevent clay particles from flocculat-
ing and making the mud too thick to circulate and to seal
the walls of the hole. The thinning agent is manufactured
from Rhizophora (mangrove) bark, quebracho (Schinopsis
lorentzii) wood extract, and oxidized coal from several
sources, including the Tongue River Formation of the
Paleocene Fort Union Group of North Dakota. Traverse et
al. (1961) recovered an estimated 600,000-4,000,000 fossil
pollen and spores per gram from these muds. This consti-
tutes a potential source of contamination in samples de-
rived from commercial drilling operations and is so recog-
nized by palynologists in the petroleum industry.
Phytoliths
By adjusting the processing procedure to exclude HF, a
wide variety of plant and animal microfossils composed of
silicate minerals often can be extracted from sedimentary
rocks. Plant material of silicate composition representing
terrestrial vegetation includes the phytoliths (Piperno,
1988), which are becoming more widely recognized as use-
ful in paleoenvironmental studies, especially for the Qua-
ternary and particularly for sediments associated with
archeological sites (Rovner, 1971). Phytoliths are micro-
122 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 4.5. Phytoliths. (A) Spherical

nodular phytolith from Maranta arundi-
naceae (9-18 um), (B) spherical smooth
phytolith from seeds ofHirtella triandra
(6-13 urn), (C) variant 1 cross-shaped
phytolith from Zea Mays, (D) seed phy-
tolith from Bursera simaruba, (E) variant
5/6 dumbell phytolith from Cenchrus
echinatus, (F) irregular pointed phytolith
from Odontocarya tamoides (36-51 x
36-46 urn). Photographs courtesy of Do-
lores Piperno (see Piperno, 1988).

scopic particles of hydrated silica produced in the cells of
many vascular plants. They vary in morphology among dif-
ferent plant groups (Fig. 4.5), are released upon decay of
the parent plant into soils, and eventually become part of
the sedimentary record. Phytoliths are common in the cells
of grasses and other plants and have been used to study
shifts in the North American deciduous forest-grassland
ecotone during the Holocene; to identify savanna grasses
in lake sediments from East Africa; to confirm cold paramo
(alpine tundra) in pollen diagrams from the highlands of
Colombia; and to document wet-dry cycles in the
Holocene of Central America from aeolian particles (in-
cluding phytoliths) recovered from DSDP and ODP cores.
[See Piperno (1988) for references.) Phytoliths have been
found in fossil plants of Miocene age in North America
(Smiley and Huggins, 1981; Thomasson, 1983, 1984;
Thomasson et al., 1986), but in kinds and abundance they
are best represented in sediments of Quaternary age.
The study of fossil pollen, spores, and phytoliths consti-
tutes one method for tracing the history of North American
vegetation. The technique has inherent strengths and limi-
tations, and a balanced appreciation of both is necessary to
most effectively use the technique as a paleoecological re-
search method.
PALEOBOTANY (PLANT MEGAFOSSILS)
The factors that determine the representation of megafos-
sils in the geologic record are different from those that de-
termine the representation of microfossils. In the Late
Miocene Succor Creek flora of southeastern Oregon, ap-
proximately 26% of the taxa were found only as megafos-
sils or as microfossils. Chaney (1959) presents quantitative
data from a study of plant remains from 19 species accu-
mulating in 42 pools along Redwood Creek in Muir Woods

National Monument, California. The most important factor
that controlled entrance into the fossilization process was
distance from the depositional basin. The four species
comprising 91.26% of the leaves and fruits grew an average
of 5.25 ft from the pools. The next group of four species,
making up 8.34% of the specimens, grew at an average dis-
tance of 25.25 ft. Even the 11 fewest species were all
within 50 ft.
The second most important factor was durability of the
organ entering the pools. The leaves of Rhododendron and
Corylus rostrata var. californica (hazel), which have a deli-
cate spongy mesophyll, were abundant in December soon
after the primary leaf fall, but nearly all had disintegrated
by March. Both are rare in most Tertiary floras in the Amer-
ican west even when their associates are common to abun-
dant. The leaves of Acer negundo (box elder) are also deli-
cate and were not an important component of the litter
even under the parent tree. The leaves of Alnus rubra, Acer
macrophyllum, and Lithocarpus densiflora are more
durable, were abundant in the pools, and are well repre-
sented in the fossil record.
A third factor was the weight-to-area ratio of the organ.
In essence, lighter specimens that can travel farther were
usually better represented than heavier ones. Also impor-
tant were the height of the plant and its size (which deter-
mines the capacity for producing leaves and reproductive
structures). Organs from large trees made up 94.25% of the
specimens in the pools, small trees and shrubs 5.68%, and
herbs 0.04%. Leaves of deciduous plants were better repre-
sented than those of evergreens.
The Late Cretaceous and Tertiary megafossil record for
North America includes an abundance of leaf material of
midsize to tall deciduous trees and large shrubs from mesic
environments that have light, broad, durable leaves, and
grow near the basins of deposition. The representation of
these plants is sufficient to provide a basis for estimating
the relative abundance and spatial arrangement of ele-
ments in these habitats and to allow environmental recon-
structions with the modern analog and foliar physiognomy
methods. Less material is available for small understory
shrubs, herbs, and evergreens; plants with heavy leaves
and reproductive structures; those growing removed from
the depositional basin; and those growing in dry to arid
habitats where conditions are not suitable for preservation.
The fewer number and smaller size of these floras and the
underrepresentation of arid elements in larger floras pro-
duces a more restricted data base for estimating their pres-
ence, abundance, arrangement, and environmental impli-
cation in a statistically meaningful sense. This limitation is
compensated for, to some extent, by the large number of
floras available in western North America, the midconti-
nent region, and the Mississippi Embayment. Fruits and
seeds have a diagnostic morphology (Corner, 1976) and are
occasionally abundant at some localities, like the Middle
Eocene Clarno flora of Oregon, and seeds are particularly
important in Quaternary peat deposits. Consideration of
Methods, Principles, Strengths, and Limitations 123
plant microfossils in conjunction with megafossils can also
minimize some of the biases in the paleobotanical record.
Modern Analog Method
Two approaches are available to analyze fossil material for
the purpose of reconstructing paleoenvironments. One is
the modern analog method or nearest living relative (NLR)
analogy. This approach involves identifying the fossils by
comparing them to modern species having the most similar
morphology. Using leaves as an example, this comparison
is based on venation, size, shape, margins, petioles, acces-
sory structures (bracts and stipules; extrafloral nectaries,
Pemberton, 1992), and where possible, variations in these
features through study of populations of the fossil material.
The morphology and variation may be compared with that
exhibited by cleared leaves (e.g., Todzia and Keating, 1991)
and X-ray radiograms (Wing, 1992) of the presumed mod-
ern analog. Microscopic study of the leaf surface reveals
details of the venation pattern and if cuticles are preserved,
these may be examined with SEM to reveal additional mor-
phological features such as stomata and trichome types
and arrangement. A standardized set of definitions and ter-
minology was developed by Hickey (1973, 1979; see also
Dilcher, 1974; Hickey and Wolfe, 1975; Wolfe, 1989). Sur-
veys of leaf venation patterns are under way to identify fea-
tures characteristic of the lineage, rather than just ones that
are the most obvious (e.g., Klucking, 1986; Roth, 1992). In
the older literature identifications were often made on the
basis of limited, fragmentary, poorly preserved specimens;
and the level of accuracy has proven to be less than spec-
tacular. The fossil leaf Carya bendirei has been described
as four different species of Salix; two species of Carya,
Rhus, and Prunus; and as Magnolia, Cassia, Celastrus, Aes-
culus, Hicoria, Juglans, Ptelea, Quercus, and Arbutus. The
old procedure of "leaf matching" has been replaced by a
critical examination of populations of well-preserved spec-
imens utilizing more sophisticated techniques, with a cor-
responding increase in the accuracy of the identifications.
Once the specimens have been identified, the altitudi-
nal and meterological conditions within the range of the
modern species are recorded and the procedure continued
for all elements of the fossil flora. Before paleoenviron-
ments can be reconstructed, however, the assemblage must
meet several criteria. First, the flora must be moderately
large in terms of the number of species present. This is de-
fined in relation to the size of other floras in the region; and
for North America, where extensive Late Cretaceous and
Cenozoic floras exist, moderately large would mean a min-
imum of about 30 species. Second, there must be a consis-
tency in climatic requirements among several, but not nec-
essarily all, members of the assemblage. In essence, this
means that paleoenvironments are estimated on the basis
of groups of species and not on individual "key" compo-
nents. Third, the ecological heterogeneity of the flora must
be accommodated through a reconstructed paleophysiogra-
124 Late Cretaceous and Cenozoic History of North American Vegetation
phy consistent with geological evidence from the region.
To properly evaluate the ecological import of the proce-
dure, it is important to note that at this point there have
been no definitive paleoclimatic reconstructions. Rather, a
preliminary model of climate and physiography has been
established for the flora. These preliminary reconstructions
must be assessed within the context of results from inde-
pendent lines of evidence. Other floras in the region, fossil
faunas, and global paleotemperature and sea-level curves
provide estimates of contemporaneous conditions and
long-term trends within which the preliminary paleoenvi-
ronmental model can be evaluated. For North American
floras at least some ancillary data are usually available. If
such data are inconsistent with the botanical evidence
(e.g., inconsistency in estimates derived from fossil mam-
malian faunas and the floras), attempts are made to provide
a plausible explanation. However, for the Late Cretaceous
and Cenozoic paleontological record of North America
such situations are rare, and a hallmark of most current in-
vestigations is an overall consistency between results de-
rived from the several lines of inquiry.
The modern analog method assumes that morphological
similarity of isolated plant parts is indicative of taxonomic
and ecological similarities to the extent that when the en-
tire assemblage of medium to large sized fossil floras with
at least moderate diversity are involved, enough of the ele-
ments will be comparable to make generalized reconstruc-
tions possible. Within limits, the assumption is valid, but
there are obvious shortcomings inherent in the method.
Many plant fossils resemble modern genera back to about
the Middle Eocene, beyond which an increasing number
represent extinct taxa. Therefore, the time interval for
which the modern analog method can be used is limited
mostly to the Middle Eocene to the present. There is also
the possibility of new forms evolving that have different
ecological parameters but retain the same or similar mor-
phological features as the parents among those parts likely
to be present in the fossil record (Chapter 3). Another com-
plication is the virtual certainty of some change in the
range and composition of plant formations and associa-
tions after each period of climatic reshuffling. As men-
tioned previously, where an organism does occur is not al-
ways the same as where it can occur, simply because it may
not have had sufficient time or opportunity to expand to its
maximum geographical and ecological range. Vegetational
histories and paleoenvironmental reconstructions based
exclusively on the morphology, distribution, and ecologi-
cal parameters of presumed modern analogs tend to ob-
scure the uniqueness of past vegetation types and dampen
the curve of vegetation dynamics and past environmental
change.
Foliar Physiognomy
The second approach to the reconstruction of paleocli-
mates is through the method of foliar physiognomy. It is a
technique that can be used independently and, after the
Middle Eocene, in conjunction with the modern analog
method for studying leaf floras.
Physiognomy is the external morphology of a structure,
organism, or community. With reference to leaves, and
specifically to those features most relevant to fossil mate-
rial, this morphology includes size, shape, texture, pres-
ence or absence of special structures (e.g., drip tips), and
characteristics of the leaf margin. Leaf size is described by
the Raunkiaer (1934) system as modified by Webb (1959)
and Wolfe (1993).2 It includes several size classes defined
as follows:
Leptophyll: maximum size 0.25 cm2
Nanophyll: 2.25 cm2
Microphyll: 20.25 cm2
Notophyll: 45 cm2
Mesophyll: 182.25 cm2
Macrophyll: 1640.2 cm2
Megaphyll: no maximum size
Leaf shape includes linear, lanceolate (Fig. 4.6A), ellipti-
cal (Fig. 4.6B), ovate, rotund, spatulate, and cuneiform
(wedge shaped; for a more complete listing of leaf forms
and terminology, see Rickey, 1973, 1979; Radford, 1986;
Stearn, 1983; Todzia and Keating, 1991; see illustrations in
Wolfe, 1993). Leaf texture is often difficult to determine
from fossil material and is commonly designated as thin,
medium, or coriaceous (sclerophyllous). In general, large
thick leaves are common among evergreens and are most
abundant in megathermal and mesothermal temperature
regimes; thinner ones in the mesophyll to notophyll size
class are usually from deciduous plants and occur in
mesothermal to microthermal climates. Drip tips (extended
acuminate; Fig. 4.6B) are a prolongation of the midvein and
associated laminar material beyond the terminus of the leaf
and aid in draining excess water from the leaf surface. In
tropical climates this is important in keeping epiphytic
algae, fungi, bryophytes, small ferns, gametophytes, and
other structures from developing to the point of causing
leaf damage and interfering with photosynthesis. The most
common types of leaf margin are entire (Fig. 4.6A), lobate,
crenate (Fig. 4.6B), dentate, serrate (Fig. 4.7), and erosus (ir-
regular).
It has long been observed that some of these features are
associated in a general way with climate (Bailey and Sin-
nott, 1915,1916; see also Givnish, 1979,1986; Givnish and
Vermeij, 1976). For example, large leaves are more preva-
lent among plants growing in the tropics and in temperate
zones in the shaded understory and along stream banks.
Smaller leaves are more frequent in open, climatically or
physiologically drier habitats (e.g., tundra, hot and cold
deserts). There is less obvious correlation between leaf
shape and climate. Drip tips are a common but not a uni-
versal feature of evergreen leaves in tropical understory
vegetation and are not found in plants of cold, mesic, or
arid habitats.
A leaf-size index has been developed to compare leaf
 Methods, Principles, Strengths, and Limitations 125

Figure 4.6. Fossil leaves illustrating

size, entire/nonentire margins, and
drip tip (or extended acuminate; see
text for discussion). (A) Eugenia ameri-
cana, Green River flora (late Middle
Eocene), Colorado and Utah (MacGini-
tie, 1969, pi. 23, fig. 2, xl). (B) Celas-
trus typica, Florissant flora (latest
Eocene), Colorado (MacGinitie, 1953,
pi. 44, fig. 5, xl). Reprinted with the
permission of the Carnegie Institution
of Washington.

size of fossil and modern vegetation and between fossil flo-
ras (Wolfe and Upchurch, 1987):
4 x M C + 3 x M E + 2 x N O + l x M I -100,
2
gories are simplified to entire and nonentire. Consider the
original tabulation of Bailey and Sinnott (1915) comparing
leaf margins from temperate (east-central North America)
and tropical (Amazon lowland) habitats:

Mesophytic-Cold-Temperate
where MC is the percent of macrophyllous (or larger)
% Entire % Nonentire
species, ME is the percent of mesophyllous, NO is the per-
cent of notophyllous, and MI is the percent of microphyl- Trees 10 90
lous. A low leaf-size index (e.g., 12-20; microphyllous) Shrubs 14 86
suggests cold or subhumid (dry) climates and full sunlight; Woody 13 87
a high index (e.g., 60-75; megaphyllous) reflects warm,
moist, aseasonal climates and/or reduced (shaded) light Moist-Lowland-Tropical
levels.
Trees 90 10
Terms and values have also been assigned to tempera-
Shrubs 87 13
ture regimes (e.g., Wolfe, 1987):
Woody 88 12
Microthermal: MAT < 13°C
Mesothermal: MAT = 13-20°C These observations have been further developed by
Megathermal: MAT > 20°C Dilcher (1973), Dolph (1978, 1984), Dolph and Dilcher
The closest association between foliar physiognomy (1979), especially Wolfe (1971, 1977, 1978, 1993), and
and climate is in the type of leaf margin (LMA). For pur- Wolfe and Hopkins (1967). Table 4.1 shows the percentage
poses of estimating paleoclimates, the leaf margin cate- of entire-margined leaves from selected vegetation types
126 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 4.7. Fossil leaves illustrating comparatively

small size and serrate margins. (A) Betula vera, Trout
Creek flora (Late Miocene, Oregon; Graham, 1965, pi.
9, fig. 2, xl). (B) Ulmus speciosa, Kilgore flora (Late
Miocene), Nebraska. Reprinted from MacGinitie
(1962, pi. 10, fig. 4, xl) with the permission of the
University of California Press.

Table 4.1. Percentages of species that have entire margins in some modern floras.
Flora Percent Vegetation
Brazil, lowland 88 Tropical rain forest
Malaya 86 Tropical rain forest
Philippine Islands, 200 m 82 Tropical rain forest
Ceylon, lowland 81 Tropical rain forest
Manila 81 Tropical rain forest
East Indies 79 Tropical rain forest
Philippine Islands, 450 m 76 Tropical rain forest
West Indies 76 Tropical rain forest
Hawaii, lowland 75 Paratropical rain forest
Ceylon, upland 74 Paratropical rain forest
Philippine Islands, 700 m 72 Paratropical rain forest
Hong Kong 72 Paratropical rain forest
Hainan, lowland 70 Paratropical rain forest
Philippine Islands, 1100 m 69 Montane rain forest
Taiwan, 0-500 m 61 Paratropical rain forest
Ceylon, upland 60 Paratropical rain forest
Hawaii, upland 57 Paratropical rain forest
Hainan, upland 55 Subtropical forest
Taiwan, 500-2000 m 45 Subtropical forest
Mixed mesophytic forest, China 30 Warm temperate forest
Northern China plain 22 Deciduous oak forest
Manchuria 10 Mixed no. hardwood forest
Adapted from Dilcher (1973) and Wolfe (1977).

extending from tropical rain forests to mixed northern
hardwood forests. The range is from 88% in a tropical rain
forest from the lowlands of Brazil to 10% in a northern
hardwood forest in Manchuria. The closest correlation is
between entire-margined leaves and MAT.
Wolfe and Upchurch (1987) believe that MAT can be re-
constructed with an accuracy of ±5% if the assemblage has
30 or more species and within ±10% if 20-29 species are
present. As they and others have recognized, however, the
translation of quantitative measurements of modern leaf
features into paleoclimates is complicated. Leaf margin
types correlate with moisture and temperature, and it is
difficult to sort out their individual effects on foliar phys-
iognomy (Dilcher, 1973). Modifications of leaf texture, size,

and margin are also a reflection of various defense strate-
gies against herbivory (Brown and Lawton, 1991). The
overprinting effect of genetic control may retard the re-
sponse of leaf morphology to changes in climate to varying
degrees among different lineages. For example, members of
the mostly temperate family Tiliaceae in the tropics have
serrate leaf margins and cordate bases, while species of the
tropical Annonaceae and Lauraceae in temperate regions
maintain entire-margined leaves (Wolfe, 1993). Assem-
blages from disturbed habitats (e.g., channel and lacustrine
deposits) often contain a disproportionally high percentage
of species with nonentire-margined leaves (Burnham,
1994). If the fossil flora represents a sere (a stage or phase)
in a successional sequence leading toward climax vegeta-
tion from a series of disturbances (volcanism, flooding,
etc.), changes in leaf margin percentages can be incorrectly
ascribed to climatic trends. Similarly, taphonomic (deposi-
tional) factors3 can select for different leaf sizes (Wolfe and
Upchurch, 1987). For example, in tracing regional vegeta-
tional histories in the American west, it is important to rec-
ognize that many Early Eocene floras are preserved in flu-
vial (stream) or paludal (swamp) sediments (e.g., the
Wilcox), while extensive Late Eocene floras (Florissant,
Green River) are from lacustrine (lake) beds. Among other
differences, streamside vegetation tends to have larger leaf
size due to shading and high humidity than vegetation
growing on exposed slopes.
After a major climatic change there is a lag time between
the change and the introduction of species compatible with
the altered conditions, the establishment of climax vegeta-
tion, and especially in the appearance of evolutionarily
modified leaf types from in situ or adjacent progenitors.
This affects the modern region and vegetation type selected
as the standard for a particular leaf margin percentage -
climate relationship and, hence, paleoclimatic reconstruc-
tions based on leaf physiognomy. Wolfe (1981) believes the
deciduous forest formation of eastern North America is too
depauperate for calibrating leaf margin percentages to cli-
mate due to the elimination of tropical forms after Late
Eocene times and to increasing isolation from western Eu-
ropean floras after about the Middle Eocene. Also, some
larger leaf types potentially present in a deciduous forest
are absent because of Arctic cold fronts that began moving
across eastern North America in the Late Neogene. Instead,
the mesic vegetation of eastern Asia was selected as the
standard for relating leaf margin percentages to climates. By
that standard, an increase of 3% entire-margined species
equals an increase of 1°C in MAT, and an entire leaf margin
percentage of about 60% equals the 20°C isotherm (Up-
church and Wolfe, 1987; Wolfe, 1979; Wolfe and Upchurch,
1987). Dolph (1990; see also Dolph, 1979; Dolph and
Dilcher, 1979) believes that cold fronts and monsoon fluctu-
ations have affected the recent vegetation of Asia. For Late
Cretaceous analyses Wolfe and Upchurch (1987) selected a
southern hemisphere scale because of the rarity of decidu-
ous angiosperms and the prominence of conifers in both
Methods, Principles, Strengths, and Limitations 127
floras. Using the southern hemisphere scale, an increase of
4% of entire-margined species equals an increase of ~1°C in
MAT, and an entire leaf margin percentage of ~68-70%
equals the 20°C isotherm. There is now a question as to
whether the modern southern hemisphere forests are a suit-
able analog for calibrating northern hemisphere Cretaceous
leaf margin percentages.
The use of formulas and the assignment of specific cli-
matic values to leaf size and leaf margin percentages may
imply that a high degree of precision is possible by this
method. In fact, a trend was developing to assign increas-
ingly exact climatic parameters to percentages of entire-
margined leaves. The lack of precise correspondence be-
tween the leaf composition in modern accumulating sedi-
ments compared to the surrounding vegetation (e.g., in
lianas from an Indiana lake; Dolph, 1984) and other data
suggest that single-variate leaf physiognomy alone pro-
vides a useful, albeit general, estimate of climate. Boyd
(1990, 1994) believes that two different vegetation types
can usually be inferred from most leaf margin ratios, "indi-
cating the use of leaf-margin data for determining paleocli-
mates is not as precise as has been claimed" (1990, p. 194).
Greenwood (1991, 1992) has found the strongest and most
consistent climatic signal in leaf assemblages emerges only
when taphonomic factors are adequately considered and
when these signals are enhanced by multivariate statistical
analysis of a large variety of community types analyzed
from several floras. As noted by Wolfe (1993) much of the
poor correlation between leaf physiognomy and climate
can be attributed to single character analyses (e.g., leaf mar-
gin) that have not been subjected to statistical tests. When
multiple characters are scored and analyzed by multivari-
ate analysis (Kovach, 1989; particularly correspondence
analysis), considerably better correlation is achieved.
Clearly, a synergistic approach is essential in paleo-
ecology (Wing, 1987). Nonetheless, a ninefold difference
in entire leaf margin percentages between the mesic and
tropical climates as listed in Table 4.1 clearly documented
foliar physiognomy as potentially an important methodol-
ogy for assigning quantitative values to Late Cretaceous
and Cenozoic climates and climatic change. In particular,
it can provide a way of estimating paleoclimates that is
independent of taxonomic identification or the assump-
tion that the ecological requirements of species have re-
mained constant over millions of years. It can further
allow estimates of paleoclimate to be made from assem-
blages older than Middle Eocene where application of the
modern analog method becomes increasing limited. Re-
quired was more sophisticated statistical analyses to
achieve closer correspondence between leaf margin per-
centages and climate.
The statistical program CLAMP (Wolfe, 1990, 1993,
1994) better quantifies the relationship between leaf charac-
ters and climate. The program utilizes 29 character states of
modern leaves from 25 to 30 species from 106 sites in North
America, the Caribbean region, and Japan; Canonical Corre-
128 Late Cretaceous and Cenozoic History of North American Vegetation
spondence Analysis relates the samples to one another, to
the character states, and to selected climatic parameters
with a moderate to high degree of accuracy (Wolfe et al.,
1996). A similar tabulation is made of leaf characters in fos-
sil floras and correspondence analysis, together with multi-
ple regression tests, provide an estimate of the climate
(mainly temperature) for the fossil assemblages with an
error range of ~2°C. In its present stage of development,
Basinger et al. (1994) believe the program underestimates
temperature and precipitation. In general, values of MAT
from CLAMP are lower than those from LMA (Wolfe, 1994).
Nonetheless, the development of LMA and CLAMP are in-
novative and important contributions to paleoecology, and
their continuing refinement and broader application will
add greater quantitative consistency to estimates of past cli-
mates. As noted by Wolfe and Hopkins, "[L]eaf-margin
analyses of large, diversified fossil floras used in conjunc-
tion with consideration of their floristic composition, can
yield reliable indications of climatic change" (1967, p. 68).
An application of fossil plant data that utilizes LMA and
CLAMP is in estimating paleoelevations. This has tradition-
ally been done by compiling the present-day altitudinal
range of presumed modern analogs and proposing a paleo-
physiography sufficient to accommodate the upland ele-
ments. Considerable latitude is possible in such estimates,
depending primarily on what MAT value and annual tem-
perature range is assumed for the assemblage. Other vari-
ables include slope and exposure, precipitation (annual
amounts and seasonal distribution), and, for progressively
older floras, changes in ecological tolerances and the possi-
bility that not all analogs are presently growing throughout
their potential altitudinal range. A new approach is to esti-
mate the MAT of a coastal or lowland fossil flora and a co-
eval upland fossil flora from the same physiographic region,
then use an assumed lapse rate (e.g., 5°C/km) to calculate
the paleoelevation of the upland assemblage (Chapter 6).
These results depend on the accuracy of the MAT and lapse
rate estimates. LMA and CLAMP are being used to provide
better approximation of the MATs.
Stomatal Analysis
A recent innovation based on stomatal features has pro-
vided another paleobotanical method for estimating paleo-
climates. Researchers have found that among modern
plants the number of stomata decreases with increases in
CO2 concentration (Beerling, 1994; Beerling and Chaloner,
1994). This has been demonstrated from herbarium mate-
rial of species collected over a 200-year period. Human-
induced increases in CO2 were found to correlate with a
40% reduction in stomata density in European forest trees,
and the trend was confirmed by laboratory experiments
using varying CO2 concentrations. Specimens of Quercus
pseudocastanea, the fossil representative of the modern
European Quercus petraea (durmast oak) were then stud-
ied from the Late Miocene through the Pliocene (Van Der
Burgh et al., 1993). Stomatal indices (stomata density/
stomata density + epidermis cell density) varied between
9.5 and 11.5 in the older material to 16.2 in the Pliocene
specimens. The increase was also evident in fossil leaves
of Fagus attenuata (10.0 ± 0.8 to 14.1 ± 0.5), and both
trends are associated with a decline in MAT of 2-3°C. Fur-
ther study of herbarium material of Q. petraea representing
a 120-year period allowed correlation of the stomatal index
with known atmospheric CO2 concentrations and gave an
estimate of 280-370 ppm in the Late Neogene compared to
338 ppm in 1980. A similar relationship for the last
glacial-interglacial cycle was found in leaves of Pinus
flexilis preserved in packrat middens from the Great Basin
(Van de Water et al., 1994). As the technique is refined and
better quantified (Kurschner, 1997), one more biosensor
will be available for estimating ancient CO2 concentration
and paleotemperature by calculating the stomatal index
from fossil leaf material.
Dendrochronology
Dendrochronology is the study of the width, density, and
isotopic composition of tree rings; it is used for recon-
structing past climates (Cook and Kairiukstis, 1989; Filion
et al., 1986; Fritts, 1966; Haugen, 1967; Hughes et al., 1982)
and detecting and dating events that leave signals in the
patterns of secondary wood development (Wiles et al.,
1996). These events include fires, epidemics, anthropolog-
ical activities (Baillie, 1982), hydrologic cycles, faulting
(Page, 1970), ENSO events (Lough and Fritts, 1990; Swet-
nam and Betancourt, 1990), and landslides and flooding
associated with seismic events (Atwater and Moore, 1992;
Jacoby et al., 1988,1992,1995; Sheppard and Jacoby, 1989)
and volcanic eruptions (LaMarche and Hirschboeck, 1984;
Lough and Fritts, 1987; Oswalt, 1957).
Growth rings are formed by the vascular cambium that
is activated in the stems of temperate trees and shrubs in
the early spring and that continues to produce secondary
wood (xylem) until late fall. Spring or early wood is
formed during the favorable growing conditions of mois-
ture, temperature, nutrient levels, and increasing day
length. The resulting tracheids and vessel elements are
comparatively large in diameter and form a band that is
light colored in appearance. Summer or late wood devel-
ops under progressively less favorable conditions, the cells
are typically smaller, and the band has a darker appear-
ance. Growth rings are evident in most conifers and are en-
hanced in many woody dicotyledons by larger diameter
vessels concentrated at the beginning of each ring (ring-
porus condition). A series of rings with broad spring wood
indicates that good growing conditions extended well into
the summer months, while narrowing bands of spring
wood across the section reflect deteriorating conditions of
rainfall and/or temperature. However, the width of a
growth ring is a function of several factors, including age,

broad genetic control, a genetically influenced "memory"
of past patterns, the local hydrologic cycle, and climate,
particularly moisture and temperature. A considerable part
of recent research in dendrochronology has been to de-
velop sampling procedures, models, and statistical tech-
niques designed to disentangle the various factors deter-
mining the width of growth rings.
The particular component of climate that can be mea-
sured through tree-ring analysis is, in part, a function of
the habitat in which the tree grows. In subpolar and alpine
regions temperature is the limiting factor, and growth rings
in trees selected from this environment will primarily re-
flect changes in temperature. In the arid southwest where
water is the limiting factor, the width of the spring wood
can either be a function of the local hydrologic cycle if the
trees are selected from floodplains or of precipitation if the
trees are selected from exposed sites with shallow soils
away from the fluctuating water levels of the floodplain.
Sampling procedures involve taking two cores per tree
from a minimum of 10 trees per site. This allows patterns
to be compared among trees and corrections made for miss-
ing and false or multiple rings. This process is called cross-
dating, and may encompass stands hundreds of kilometers
apart. Another necessary procedure is standardization. As
trees age, the width of the rings becomes narrower. It is
necessary to apply a factor to correct for this and other ex-
ternal influences on growth, such as closing or opening of
the canopy from wind damage, pathogens, and other fac-
tors. This is accomplished by calculating the mean width
of rings in young and older portions of the stem, represent-
ing periods of comparable growing conditions, among sev-
eral trees at a site. The exponential is then applied to the
measurements of each ring width. A mean site chronology
is established by combining the individual standardized
chronologies.
Calibration of the series is necessary to allow associa-
tion of a given width with the values of a particular cli-
matic parameter (viz., annual rainfall). This is done by
comparing each corrected ring width with known metero-
logical data. The next step is verification, whereby metero-
logical conditions are estimated from ring widths in other
sections of the tree and then compared with actual weather
information for that period.
With a series of tree-ring chronologies with standard-
ized ring-width indices established from which precip-
itation and temperature values can be read, it is possible
to estimate climatic changes by utilizing ancient, slow-
growing trees common to the region. The oldest records are
from the arid southwest, based on the early works of as-
tronomer A. E. Douglas (1909), his student Edmund Schul-
man (1956), and H. C. Fritts (1976, 1991) at the Laboratory
of Tree-Ring Research at the University of Arizona. By
using living Pin us aristata (bristlecone pine, 4600 years
old; Figs. 4.8, 4.9) and other trees, augmented by timbers
and charcoal at archeological sites, and sub fossils from
caves, playa lakes, and bogs, it has been possible to estab-
Methods, Principles, Strengths, and Limitations 129
lish a continuous chronology back 8800 years. Other labo-
ratories at the Universities of Arkansas, Nebraska, and
Washington, and at the Lamont-Doherty Earth Observatory
(Columbia University), Oak Ridge National Laboratory, and
the U.S. Geological Survey at Reston, Virginia, have estab-
lished shorter chronologies for different parts of North
America. The oldest living trees in eastern North America
are 1700-year-old Taxodium distichum (bald cypress)
growing along the Black River in North Carolina (Stahle et
al., 1988). An International Tree-Ring Data Bank is based at
the University of Arizona as part of a developing interna-
tional tree-ring network (Stockton et al., 1985).
Measurements of ring widths utilizing appropriate sta-
tistical analyses is one procedure for estimating past cli-
mates. A second area of promising research is tree-ring iso-
tope chemistry (Stuiver and Burk, 1985). This technique is
based on the original observations by Urey (1947) that fluc-
tuations in the isotopes of organic material may reflect the
ambient temperature prevailing at the time the material
was formed. The development and application of this ob-
servation to cellulose across the secondary xylem is called
tree thermography (Libby, 1987). The determination of car-
bon ratios involves oxidizing the carbon in the sample to
CO2 and analyzing the gases with a mass spectrometer. The
relative amounts of 13C and 12C in the wood is a function of
their abundance in the atmosphere that in turn varies with
temperature: increasing temperatures result in larger
amounts of the heavier 13C and a larger 13C/12C ratio. Cur-
rent research involves determining a 13C/12C isotopic tem-
perature coefficient whereby paleotemperatures can be es-
timated from a given ratio.
Isotopes of hydrogen (deuterium to hydrogen; Feng and
Epstein, 1994) and oxygen (18O to 16O) from cellulose are
also being used to estimate paleotemperatures. The relative
amounts of these isotopes in the atmosphere varies accord-
ing to ocean-surface temperatures. In warm oceans more of
the heavier isotopes are distilled, while in a cooling ocean
less are evaporated. These variations are preserved in the
chemical composition of cellulose forming growth rings. To
initially determine whether isotope ratios reflected ambient
temperature, it was necessary to analyze the cellulose from
trees growing near sites with long meteorological records.
The longest mercury thermometer records are from Ger-
many and date from 1690. The oldest tree near the metero-
logical recording station was an oak that provided rings
from 1700 to 1950. The tree-ring isotope patterns matched
the actual recorded changes in temperature; both recorded
such climatic events as the warm period of 1710-1740, a
cold period from 1740 to 1800, and a slow warming until
the present with a peak in 1930 (Libby, 1987).
One of the strengths of dendrochronology is that each
climatic trend and event is encoded within a specific set of
annually formed growth rings and it can be precisely dated.
Also, the data are comparatively easy to obtain from any
temperate forested region. The paleoclimatic information is
not dependent on taxonomic identification or on the as-
 130 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 4.8. Pinus aristata
(bristlecone pine), Arizona.
Photograph provided by
Charles W. Stockton.
sumption of ecological consistency for a species over long
periods of geologic time (Creber and Chaloner, 1985). The
principal limitation is the relatively short time span over
which the technique can be applied. Also, some aspects of
tree-ring analysis cannot be applied to trees growing in
tropical environments because of the lack or poor develop-
ment of growth rings (Jacoby, 1989). However, just the pres-
ence or absence of rings in petrified wood can provide use-
ful ancillary information for assessing past climates based
on other approaches (Creber and Chaloner, 1984). Gener-
ally, the absence of distinct growth rings (Fig. 4.10) among
several woods at a locality means nonseasonal tropical cli-
mates prevailed, while their presence (Fig. 4.11) suggests
more temperate or seasonal climates. In the fossil record of
North America the former are common in the Late Creta-
ceous and Paleocene in the midlatitudes, and the latter
mostly characterize the Eocene and later times. In this con-
nection, it is important to note, however, that the presence
of growth rings in fossilized wood does not automatically
exclude the existence of tropical to subtropical paleoenvi-
ronments. Tomlinson and Craighead (1972) examined the
wood of 87 species growing in Florida south of latitude 26°
N (viz., south of Ft. Lauderdale). Well-developed growth
rings were present in 51 species, while 36 species produced
no or poorly developed rings. It is also difficult sometimes
to determine whether growth patterns in the secondary

Figure 4.9. Cross section of stem ofPinus aristata (bristlecone
pine), illustrating growth rings. Photograph provided by Charles
W. Stockton.
xylem are due to precipitation, temperature, or genetic
memory. Some ecological implications of wood features are
summarized in Fig. 4.12, and other aspects of wood anat-
omy and environment are discussed by Carlquist (1975,
1977,1988) and Wheeler and Baas (1991,1993).
Recent refinements in standardizing tree-ring chronolo-
gies, the development and application of statistical meth-
ods of analysis, and the emerging techniques of isotope
chemistry have established dendrochronology as an in-
creasingly precise and useful technique for the study of
vegetational history and paleoenvironments.
Packrat Middens
Part way up we came to a high cliff and in its face were
niches . . . and in some of them we found balls of a glis-
tening substance looking like pieces of variegated candy
. . . it was evidently food of some sort, and we found it
sweet but sickish, and those who were hungry, making
a good meal of it, were a little troubled with nausea af-
terwards. From the diary of a lost prospector in the Gold
Rush of 1849. (Jaroff, 1992)
Another method for studying the recent history of vegeta-
tion is through analysis of packrat (or woodrat) middens.
These rodents belong to the genus Neotoma (Fig. 4.13);
there are -20 species living from sea level to 3350-m eleva-
tion across North America and ranging from the latitudes
of British Columbia (-50° N) to Guatemala (-16° N). They
forage for food and construct houses or dens from nearby
vegetation, usually within a radius of less than 50 m from
the site. The houses are frequently built in cliff recesses,
rock shelters (Fig. 4.14), and caves. The middens also serve
Methods, Principles, Strengths, and Limitations 131
as urinal perches that become hardened over time (Fig.
4.15) and, in arid protected environments, they are pre-
served as consolidated layers of plant debris that reflect the
local surrounding vegetation throughout the time the mid-
dens were being constructed. Because preservation is fa-
vored by dryness, the analysis of packrat middens as a
method of paleovegetation analysis is confined to the arid
southwest, even though the range of the organism is con-
siderably greater. The principal species in the Chihuahuan
Desert is the white-throated packrat (N. albigula; Fig. 4.13).
In the Mohave Desert it is the desert packrat (N. lepida),
and on the Colorado Plateau and elsewhere in the Great
Basin it is the bushy-tailed packrat (N. cinerea).
The oldest known middens are more than 50 Ky from
central Nevada (Spaulding, 1985) and 51 Ky from western
Arizona (14C finite date; K. L. Cole, unpublished data), but
most date from the late glacial through the Holocene. Sites
from the Chihuahuan Desert (Van Devender, 1990) have
yielded about 480 plant taxa, most of which have been
identified to species on the basis of megafossil remains
(Fig. 4.16). The species found in the middens represent
mostly escarpment vegetation because the middens that
survive are those constructed in protected cliff-side cave
sites and, as noted, the packrats range only -50 m from the
middens. The high organic content of the structures read-
ily allows radiocarbon dating and about 1000 dates are
now available from middens in the southwest. The relative
abundance of identified megafossils from each level consti-
tutes a spectrum, which provides a general picture of the
surrounding vegetation for one point in time. The spectra
can be arranged in a vertical sequence through the midden
to constitute a profile for each taxon, and the profiles col-
 Figure 4.10. Fossil wood of Paraphyllanthoxylon abbottii,
Paleocene, Big Bend National Park, Texas. Growth rings are
indistinct to absent. Photograph courtesy of Elisabeth
Wheeler (see Wheeler, 1991).

Figure 4.11. Fossil wood of Magnolia longiradiata, middle

Eocene Clarno Formation, Oregon. Growth rings are dis-
tinct. Photograph courtesy of Elisabeth Wheeler (see Scott
and Wheeler, 1982).
 Methods, Principles, Strengths, and Limitations 133

Cool Temperate or Montane Habitat

— decrease in vessel element length
— decrease in vessel diameter
— increase in vessel frequency
and in temperate seasonal climate
increase in helical thickenings
Xeric Habitat i

— decrease in vessel element length

— decrease in vessel diameter Tropical Lowland Habitat
— increase in vessel frequency — increase in vessel diameter
— increase in vessel dimorphism — decrease in vessel frequency
— increase in wall thickness — increase in simple perforations
— increase in simple perforations — decrease in scalariform perforation
and decrease in scaJariform perforation plates and bar number
plates and bar number
- increase in vessel groupings Tropical, Mesic
Woody Shrubs
or Trees
(Middle Elevations)
High incidence (~20%) of many-barred
scnlariform perforation plates in long,
narrow, angular, thin-walled vessel
elements
Figure 4.12. Trends in vessel element specialization reflecting different habitats. Reprinted from Wheeler and
Baas (1991; adapted from Dickison, 1989) with the permission of the International Association of Wood
Anatomists.

pellets, and the size decreases during warm periods fol-

Figure 4.1 3. White-throated pack rat (Neotoma albigula),
a mostly desert-inhabiting species. Photograph courtesy
of Terry Vaughan (see Vaughan, 1990) and reprinted with
the permission of the University of Arizona Press.
lectively make a diagram that reflects changes in the vege-
tation over time (Fig. 4.17). This proxy data can be used to
infer changes in climate. The disappearance of the Anasazi
tribe from Chaco Canyon in New Mexico -1200 years B.P.
correlates with depletion of pine debris in the middens.
The inference is that the overuse of the woodland for con-
struction and fuel may have contributed to erosion and
loss of the land for farming (Betancourt and Van Devender,
1981). In addition, body size can be estimated from fecal
lowing Bergmann's Rule (Smith et al., 1995). Fluctuations
in body size of N. cinerea from the Great Basin and Col-
orado Plateau closely track temperature changes simulated
by the CCM at the NCAR.
A recent innovation is analysis of pollen and spores pre-
served in the acid environments provided by the middens.
The plant microfossils mostly represent anemophilous
pollen and spores blown into the caves by wind. Pollen
and spore concentrations in middens have been reported
as ranging from 17,000 to 197,000 grains/g by Thompson
(1985), 4870 to 629,508 grains/g by Davis and Anderson
(1987), and 23,003 to 533,500 grains/g by Anderson and
Van Devender (1991). Other sources of microfossils are
flowers, and pollen and spores adhering to plant surfaces
brought into the middens as food or nesting material.
Minor amounts may be brought in on the pelts and re-
moved by preening, while others are preserved in fecal pel-
lets. Pollen and spore concentrations in the latter range
from about 12,350 to 408,350 grains/g and include zo-
ophilous species. However, they may reflect dietary prefer-
ences more than a random sampling of the surrounding
vegetation.
The procedure for isolating pollen and spores from the
middens involves dissolving the binding urine matrix in
water and treating the residue with 10% KOH and hot
water washes that dissolve cuticular and cellulosic debris
and clear the darkened specimens. Hydrofluoric acid re-
moves silicates, and acetolysis (treatment with a mixture of
 Figure 4.14. Midden of bushy-tailed pack rat (Neotoma cinerea), composed mostly of
saltbrush (Atriplex canescens), Navajo County, Arizona. Photograph courtesy of Terry
Vaughan (see Vaughan, 1990; photograph in original publication is upside down) and
reprinted with the permission of the University of Arizona Press.

Figure 4.15. Indurated pack rat midden, Eleana Range, Nevada, 17.1-10.6 ka. Photograph courtesy of W. Geof-
frey Spaulding (see Spaulding et al, 1990) and reprinted with the permission of the University of Arizona
Press.

134
Figure 4.16. Plant megafossils from pack rat midden, Maravillas Canyon, Texas. Scale
division in millimeters. (A) Papershell pinon needles, 21 ka; (B) Thompson/beaked
yucca (Yucca thompsoniana/rostrata) leaf tips, 21 ka; (C) juniper twig, 21 ka; (D) shrub
oak leaf, 20.6 ka; (E) mock pennyroyal (Hedeoma plication) leaf and fruit, 20.6 ka; (F)
rock daisy (Perityle sp.) achene, 20.6 ka; (G) goldeneye (Viguiera cf. dentata), 16.2 ka;
(H) winter fat (Ceratoides lanata) leaf, 16.2 ka; (I) desert olive (Forestiem sp.) seed, 16.6
ka; (J) gray daisy (Bahia absinthifolia] achene, 16.6 ka; (K) false nightshade (Chamae-
saracha sp.) seed, 6 ka; (L) blind prickly pear (Opuntia sp.) seed, 5.1 ka; (M) prickly pear
cactus (Opuntia leptocaulis] stem, 5.1 ka; (N) roemer catclaw (Mimosa sp.) twig, 5.1 ka;
(O) resurrection plant (Selaginella) stem, 5.1 ka. Reprinted from Van Devender (1990)
with the permission of the University of Arizona Press.

135
Figure 4.1 7. Megafossil profile from pack rat midden, Maravillas Canyon, Texas. Reprinted from Van Devender (1990) with the permission of the
University of Arizona Press.

9 parts of acetic anhydride to 1 part of concentrated sulfu-
ric acid) oxidizes other organic debris. The residue is
mixed with a mounting medium, such as glycerine jelly or
silicone oil, and mounted on slides for examination.
In addition to spores, pollen, and megafossils, the plant
debris comprising middens also preserves changes in
13 12
C/ C ratios in atmospheric CO2. It has been found, for ex-
ample, that the pattern of isotopically lighter atmospheric
CO2 during the last ice age, compared to interglacial periods,
is preserved in the midden material (Marino et al., 1992).
One of the principal contributions of the study of packrat
middens is the provision of a record of past vegetation from
regions where other types of paleovegetation analysis can-
not be made. For example, Anderson and Van Devender
(1991) present a pollen record from the Waterman Moun-
tains of southern Arizona. This is a hot, dry environment
where plant mega- and microfossils preserved in middens
are the only extensive source of data on the late-glacial and
Holocene history of the vegetation. Another advantage is
that the megafossils often can be identified to species. Mid-
den material also provides abundant organic debris suitable
for 14C dating and establishing high-resolution chronologies.
A limitation of the technique is that it is applicable to a
limited geographic region (the arid southwest) and to a rela-
tively brief segment of time (mostly since 35 Kya). However,
a 35 Ky record gives a detailed picture of vegetational
change across a major glacial-interglacial boundary. A
source of difficulty in interpretation is that the longer se-
quences are reconstructed by amalgamating results from
several middens preserved in caves from regions that often
differ slightly in topography, lithology, and vegetation. Be-
havioral differences among the species could result in some
preferential selection of material for food and nest con-
struction; and where composite sequences are built up from
individual middens, this selectivity might be reflected in
the analysis. There is often a lack of precise stratigraphic
control between sites, and assemblages of different ages are
occasionally mixed at a site. However, development of tan-
dem accelerator mass spectrometry (TAMS), whereby par-
ticles as small as individual seeds can be dated, makes this
less of a problem (Van Devender et al., 1985). In this tech-
nique 14C atoms are counted directly, rather than having to
count the rare B particles emitted.
There are also the usual cadre of strengths and weak-
nesses inherent in the microfossil versus megafossil ap-
proaches. As noted previously, pollen grains and spores
can mostly be identified to genus, or in some cases only to
family, but they generally provide a more regional repre-
sentation of the past vegetation. The megafossils can often
be identified to genus and even to species, but they often
provide a record biased toward the local canyon or cliff-
side vegetation. These habitats can include a dispropor-
tionately large number of relict species from previous cli-
matic cycles and contain more mesic trees and shrubs than
the surrounding treeless desert. Although the relative mer-
its can be debated (Hall, 1985; Spaulding, 1990; Van De-
Methods, Principles, Strengths, and Limitations 137
vender, 1988), new and valuable information is being de-
rived from the study of megafossils and microfossils pre-
served in packrat middens (Anderson and Van Devender,
1991; Hall, 1986; Thompson, 1985).
The utilization of plant microfossils (pollen, spores, phy-
toliths), megafossils (stomatal analysis, dendrochronology,
cellulose isotope chemistry), modern analogs, and leaf
physiognomy (including newly developing computer pro-
grams and appropriate statistical tests) provides an impres-
sive arsenal of botanical techniques available for tracing
the development of vegetation through time. A balance is
now emerging between the older preoccupation with
"what is it?" and new opportunities to assess "what does it
mean?". As greater use is made of other diverse and inde-
pendent sources of context information (Chapters 2, 3), the
descriptive and causal history of North American environ-
ments and biotas, from the Cretaceous to the present, can
both be outlined in rather impressive detail. This history is
recounted in the following chapters.
References
Anderson, R. S. and T. R. Van Devender. 1991. Comparison
of pollen and macrofossils in packrat (Neotoma) mid-
dens: a chronological sequence from the Waterman
Mountains of southern Arizona, U.S.A. Rev. Palaeobot.
Palynol. 68: 1-28.
Atwater, B. E and A. L. Moore. 1992. A tsunami about 1000
years ago in Puget Sound, Washington. Science 258:
1614-1617.
Bailey, I. W. and E. W. Sinnott. 1915. A botanical index of
Cretaceous and Tertiary climates. Science 41: 831-834.
. 1916. The climatic distribution of certain types of
angiosperm leaves. Am. J. Bot. 3: 24-39.
Baillie, M. G. L. 1982. Tree-Ring Dating and Archeology.
Groom Helm, London.
Basinger, J. E, D. R. Greenwood, and T. Sweda. 1994. Early
Tertiary vegetation of Arctic Canada and its relevance
to paleoclimatic interpretation. In: M. C. Boulter and
H. C. Fisher (eds.), Cenozoic Plants and Climates of the
Arctic. NATO ASI Series 1: Global Environmental
Change. Springer-Verlag, Berlin. Vol. 27, pp. 175-198.
Beerling, D. J. 1994. Palaeo-ecophysiological studies on
Cretaceous and Tertiary fossil floras. In: M. C. Boulter
and H. C. Fisher (eds.), Cenozoic Plants and Climates
of the Arctic. NATO ASI Series 1: Global Environmen-
tal Change. Springer-Verlag, Berlin. Vol. 27, pp. 23-33.
and W. G. Chaloner. 1994. Atmospheric CO2
changes since the last glacial maximum: evidence
from the stomatal density record of fossil leaves. Rev.
Palaeobot. Palynol. 81: 11-17.
Betancourt, J. L. and T. R. Van Devender. 1981. Holocene
vegetation in Chaco Canyon, New Mexico. Science
214: 656-658.
Boyd, A. 1990. The Thyra 0 flora: toward an understanding
of the climate and vegetation during the Early Tertiary in
the high Arctic. Rev. Palaeobot. Palynol. 62:189-203.
. 1994. Some limitations in using leaf physiognomic
data as a precise method for determining paleocli-
mates with an example from the Late Cretaceous
138 Late Cretaceous and Cenozoic History of North American Vegetation
Pautut flora of west Greenland. Palaeogeogr., Palaeocli-
matol., Palaeoecol. 112: 261-278.
Brooks, J., R R. Grant, M. D. Muir, R van Gijzel, and G. Shaw
(eds.). 1971. Sporopollenin. Academic Press, New York.
Brown, V. K. and J. H. Lawton. 1991. Herbivory and the
evolution of leaf size and shape. Phil. Trans. R. Soc.
Lond. B 333: 265-272.
Burnham, R. J. 1994. Paleoecological and floristic hetero-
geneity in the plant-fossil record—an analysis based
on the Eocene of Washington. U.S. Geol. Surv. Bull.
2085B: 1-36.
Carlquist, S. 1975. Ecological Strategies of Xylem Evolu-
tion. University of California Press, Berkeley, CA.
. 1977. Ecological factors in wood evolution: a floris-
tic approach. Am. J. Bot. 64: 887-896.
1988. Comparative Wood Anatomy. Springer-
Verlag, Berlin.
Chaney, R. W 1959. Miocene floras of the Columbia
Plateau. Part I. Composition and interpretation. Car-
negie Inst. Wash. Contrib. Paleontol. 617: 1-134.
Cook, E. R. and L. A. Kairiukstis (eds.). 1989. Methods of
Dendrochronogy. Kluwer Academic Publishers, Boston.
Corner, E. J. H. 1976. The Seeds of Dicotyledons. Cam-
bridge University Press, Cambridge, U.K.
Creber, G. T and W. G. Chaloner. 1984. Climatic indications
from growth rings in fossil woods. In: P. J. Brenchley
(ed.), Fossils and Climate. Wiley, New York. pp. 49-74.
. 1985. Tree growth in the Mesozoic and Early Ter-
tiary and the reconstruction of palaeoclimates. Palaeo-
geogr., Palaeoclimatol., Palaeoecol. 52: 35-60.
Crepet, W. L., K. C. Nixon, E. M. Friis, and J. V. Freunden-
stein. 1992. Oldest fossil flowers of hamamelidaceous
affinity, from the Late Cretaceous of New Jersey. Proc.
Natl. Acad. Sci. USA 89: 8986-8989.
Davis, M. B. and K. Faegri. 1967. Translation: Forest tree
pollen in south Swedish peat bog deposits, by Lennart
von Post. Pollen et Spores 9: 375-401.
Davis, O. K. and R. S. Anderson. 1987. Pollen in packrat
(Neotoma) middens: pollen transport and the relation-
ship of pollen to vegetation. Palynology 11: 185-198.
Dickison, W. C. 1989. Steps toward the natural system of the
dicotyledons: vegetative anatomy. Aliso 12: 555-566.
Dilcher, D. L. 1963. Cuticular analysis of Eocene leaves of
Ocotea obtusifolia. Am. J. Bot. 50: 1-8.
. 1973. A paleoclimatic interpretation of the Eocene
floras of southeastern North America. In: A. Graham
(ed.), Vegetation and Vegetational History of Northern
Latin America. Elsevier Science Publishers, Amster-
dam, pp. 39-59.
1974. Approaches to the identification of an-
giosperm leaf remains. Bot. Rev. 40: 1-157.
Dolph, G. E. 1978. Variation in leaf size and margin type
with respect to climate. Cour. Forsch. Inst. Sencken-
berg 30: 153-158.
. 1979. Variation in leaf margin with respect to cli-
mate in Costa Rica. Bull. Torrey Bot. Club 106: 104-
109.
. 1984. Leaf form of the woody plants of Indiana as
related to environment. In: N. S. Margaris, M. Ari-
anoustou-Farragitaki, and W. C. Oechel (eds.), Being
Alive on Land, Tasks for Vegetation Science. Dr. W.
Junk, Publisher, The Hague. Vol. 13, pp. 51-61.
. 1990. A critique of the theoretical basis of leaf mar-
gin analysis. Proc. Indiana Acad. Sci. 99: 1-10.
and D. L. Dilcher. 1979. Foliar physiognomy as an
aid in determining paleoclimate. Palaeontograph. Abt.
B, 170: 151-172.
Douglas, A. E. 1909. Weather cycles in the growth of big
trees. Monthly Weather Review 37: 225-237.
Efremov, I. A. 1940. Taphonomy: a new branch of paleon-
tology. Pan Am. Geol. 74: 81-93.
Faegri, K., R E. Kaland, and K. Krzywinski. 1989. Textbook
of Pollen Analysis, 4th edition. Wiley, New York.
Farley, M. B. 1990. Vegetation distribution across the Early
Eocene depositional landscape from palynological
analysis. Palaeogeogr., Palaeoclimatol., Palaeoecol. 79:
11-27.
Feng, X. and S. Epstein. 1994. Climatic implications of an
8000-year hydrogen isotope time series from bristle-
cone pine trees. Science 265: 1079-1081.
Filion, L., S. Payette, L. Gauthier, and Y. Boutin. 1986. Light
rings in subarctic conifers as a dendrochronological
tool. Quatern. Res. 26: 272-279.
Frederiksen, N. O. 1989. Late Cretaceous and Tertiary flo-
ras, vegetation, and paleoclimates of New England.
Rhodora 91: 25-48.
Fritts, H. C. 1966. Growth-rings of trees: their correlation
with climate. Science 154: 973-979.
. 1976. Tree Rings and Climate. Academic Press,
New York.
-. 1991. Reconstructing Large-Scale Climatic Patterns
from Tree-Ring Data: A Diagnostic Analysis. Univer-
sity of Arizona Press, Tucson, AZ.
Gastaldo, R. A. 1988. A conspectus of phytotaphonomy. In:
W. A. DiMichele and S. L. Wing (eds.), Methods and
Applications of Plant Paleoecology. Paleontological
Society, Knoxville, TN, Special Publ. 3. pp. 14-28.
Givnish, T. J. 1979. On the adaptive significance of leaf
form. In: O. T. Solbrig, S. Jain, G. B. Johnson, and P. H.
Raven (eds.), Topics in Plant Population Biology. Co-
lumbia University Press, New York. pp. 375-407.
. 1986. On the Ecology of Plant Form and Function.
Cambridge University Press, Cambridge, U.K.
and G. J. Vermeij. 1976. Sizes and shapes of liane
leaves. Am. Naturalist 110: 743-770.
Graham, A. 1965. The Sucker Creek and Trout Creek
Miocene Floras of Southeastern Oregon. Kent State
University, Kent, OH. Research Ser. IX.
Gray, J. (Coordinator). 1965. Techniques in palynology. In:
B. Kummel and D. Raup (eds.), Handbook of Paleonto-
logical Techniques. W. H. Freeman and Co., San Fran-
cisco, pp. 471-706.
Greenwood, D. R. 1991. The taphonomy of plant macrofos-
sils. In: S. K. Donovan (ed.), The Processes of Fossiliza-
tion. Belhaven Press, London, pp. 141-169.
. 1992. Taphonomic constraints on foliar physiog-
nomic interpretations of Late Cretaceous and Tertiary
palaeoclimates. Rev. Palaeobot. Palynol. 71: 149-190.
Hall, S. A. 1985. Quaternary pollen analysis and vegeta-
tional history of the southwest. In: V M. Bryant, Jr. and
R. G. Holloway (eds.), Pollen Records of Late-Quater-
nary North American Sediments. American Associa-
tion of Stratigraphic Palynologists Foundation, Dallas,
TX. pp. 95-123.
. 1986. Plant macrofossils from wood rat middens:
vegetation or flora? American Quaternary Association
9th Biennial Meeting, Champaign-Urbana, June 2-4,
1986. Program and Abstracts, p. 136.

Haugen, R. K. 1967. Tree ring indices: a circumpolar com-
parison. Science 158: 773-775.
Herendeen, P. S. 1991. Lauraceous wood from the mid-
Cretaceous Potomac Group of eastern North America:
Paraphyllanthoxylon marylandense sp. nov. Rev.
Palaeobot. Palynol. 69: 277-290.
Hickey, L. J. 1973. Classification of the architecture of di-
cotyledonous leaves. Am. J. Bot. 60: 17-33.
. 1979. A revised classification of the architecture of
dicotyledonous leaves. In: C. R. Metcalfe and L. Chalk
(eds.), Anatomy of the Dicotyledons. Clarendon Press,
Oxford, U.K. Vol. I, pp. 25-39.
and J. A. Wolfe. 1975. The bases of angiosperm phy-
logeny: vegetative morphology. Ann. Missouri Bot.
Garden 62: 538-589.
Hoffmeister, W. S. 1959. Palynology's first 10 years as an
aid to finding oil. Oil Gas J., August 17.
Hughes, M. K., P. M. Kelley, J. R. Pilcher, and V C.
Lamarche, Jr. (eds.). 1982. Climate from Tree Rings.
Cambridge University Press, Cambridge, U.K.
Hyde, H. A. and D. W. Williams. 1944. Right word. Pollen
Anal. Circ. 8: 6.
Jacoby, G. C. 1989. Overview of tree-ring analysis in tropi-
cal regions, Internatl. Assoc. Wood Anat. Bull. 10: 99-
108.
, G. Carver, and W. S. Wagner. 1995. Trees and herbs
killed by an earthquake -300 yr ago at Humboldt Bay,
California. Geology 23: 77-80.
Jacoby, G. C., R R. Sheppard, and K. E. Sieh. 1988. Irregular
recurrence of large earthquakes along the San Andreas
fault: evidence from trees. Science 241: 196-199.
Jacoby, G. C., R L. Williams, and B. M. Buckley. 1992. Tree-
ring correlation between prehistoric landslides and
abrupt tectonic events in Seattle, Washington. Science
258: 1621-1623.
Jansonius, J. and D. C. McGregor (eds.). 1996. Palynology:
Principles and Applications. American Association of
Stratigraphic Palynologists Foundation, Dallas, TX.
Jaroff, L. 1992. Nature's time capsules. Time Magazine
April 6: 61.
Kapp, R. 0.1961. Modern pollen rain studies in the central
prairie of North America. Bull. Ecol. Soc. Am. 42:
110-111.
Klucking, E. R 1986. Leaf Venation Patterns. Annonaceae. J.
Cramer, Stuttgart. Vol. 1.
Kovach, W. L. 1989. Comparisons of multivariate analytical
techniques for use in pre-Quaternary plant paleoecol-
ogy. Rev. Palaeobot. Palynol. 60: 255-282.
Kremp, G. O. W. 1983. Fourth supplement, annotated refer-
ences, Precambrian to Tertiary palynological litera-
ture. Paleo Data Banks 20: 1-68.
and J. G. Methvin. 1968. Taxonomic crisis in pre-
Pleistocene palynology. Oklahoma Geol. Notes 28:146-
153.
Kurschner, W. M. 1997. The anatomical diversity of recent
and fossil leaves of durmast oak (Quercus petraea
Lieblein/Q. pseudocastanea Goeppert)—implications
for their use as biosensors of palaeoatmospheric CO2
levels. Rev. Palaeobot. Palynol. 96: 1-30.
LaMarche, V. C., Jr. and K. K. Hirschboeck. 1984. Frost rings
in trees as records of major volcanic eruptions. Nature
307: 121-126.
Leopold, E. B., G. Liu, and S. Clay-Poole. 1992. Low-
biomass vegetation in the Oligocene? In: D. R. Prothero
Methods, Principles, Strengths, and Limitations 139
and W. A. Berggren (eds.), Eocene-Oligocene Climatic
and Biotic Evolution. Princeton University Press,
Princeton, NJ. pp. 399-420.
Libby, L. M. 1987. Evaluation of historic climate and pre-
diction of near-future climate from stable-isotope vari-
ations in tree rings. In: M. R. Rampino, J. E. Sanders,
W. S. Newman, and L. K. Konigsson (eds.), Climate-
History, Periodicity, and Predictability. Van Nostrand
Reinhold Co., New York. pp. 81-89.
Lough, J. M. and H. C. Fritts. 1987. An assessment of the
possible effects of volcanic eruptions on North Ameri-
can climate using tree-ring data, 1602-1900 A.D. Cli-
matic Change 10: 219-239.
. 1990. Historical aspects of El Nino/Southern Oscil-
lation-information from tree rings. In: E W. Glynn (ed.),
Global Ecological Consequences of the 1982-83 El
Nino-Southern Oscillation. Elsevier Oceanography Se-
ries. Elsevier Science Publishers, Amsterdam. Vol. 52.
MacGim'tie, H. D. 1953. Fossil plants of the Florissant Beds,
Colorado. Carnegie Inst. Wash. Contrib. Paleontol. 599.
. 1962. The Kilgore flora, a Late Miocene flora from
northern Nebraska. Univ. Calif. Publ. Geol. Sci. 35:
67-158.
. 1969. The Eocene Green River flora of northwest-
ern Colorado and northeastern Utah. Univ. Calif. Publ.
Geol. Sci. 83:1-203.
Marino, B. D., M. B. McElroy, R. J. Salawitch, and W. G.
Spaulding. 1992. Glacial-to-interglacial variations in
the carbon isotopic composition of atmospheric CO2.
Nature 357: 461-466.
Oswalt, W. H. 1957. Volcanic activity and Alaskan spruce
growth in A.D. 1783. Science 126: 928-929.
Page, R. 1970. Dating episodes of faulting from tree rings:
effects of the 1958 rupture of the Fairweather fault on
tree growth. Geol. Soc. Am. Bull. 81: 3085-3094.
Pearson, D. L. 1984. Pollen/Spore Color "Standard."
Phillips Petroleum Co., Bartlesville, OK, Exploration
Projects Section. Version 2, Privately circulated.
Pemberton, R. W. 1992. Fossil extrailoral nectaries, evi-
dence for the ant-guard antiherbivore defense in an
Oligocene Populus. Am. J. Bot. 79: 1242-1246.
Piperno, D. R. 1988. Phytolith Analysis. Academic Press,
New York.
Radford, A. E. 1986. Fundamentals of Plant Systematics.
Harper & Row, New York.
Raunkiaer, C. 1934. The Life-Forms of Plants and Statistical
Plant Geography. Oxford University Press, Oxford, U.K.
Ritchie, J. C. and S. Lichti-Federovich. 1967. Pollen disper-
sal phenomena in Arctic-subarctic Canada. Rev.
Palaeobot. Palynol. 3: 255-266.
Roth, I. 1992. Leaf Structure: Coastal Vegetation and Man-
groves of Venezuela. Gebriider Borntraeger, Berlin.
Rovner, 1.1971. Potential of opal phytoliths for use in paleo-
ecological reconstruction. Quatern. Res. 1: 343-359.
Schulman, E. 1956. Dendroclimatic changes in semiarid
America. University of Arizona Press, Tucson, AZ.
Scott, R. A. and E. A. Wheeler. 1982. Fossil woods from the
Eocene Clarno Formation of Oregon. Internatl. Assoc.
Wood Anat. Bull. 3: 135-154.
Shaw, G. 1971. The chemistry of sporopollenin. In: J.
Brooks, P. R. Grant, M. Muir, R van Gijzel, and G. Shaw
(eds.), Sporopollenin. Academic Press, New York. pp.
305-350.
Sheppard, P. R. and G. C. Jacoby. 1989. Application of tree-
140 Late Cretaceous and Cenozoic History of North American Vegetation
ring analysis to paleoseismology: two case studies. Ge-
ology 17: 226-229.
Smiley, C. J. and L. M. Huggins. 1981. Pseudofagus ida-
hoensis, N. Gen. et sp. (Fagaceae) from the Miocene
Clarkia flora of Idaho. Am. J. Dot. 68: 741-761.
Smith, E A., J. L. Betancourt, and J. H. Brown. 1995. Evolu-
tion of body size in the woodrat over the past 25,000
years of climate change. Science 270: 2012-2014.
Spaulding, W. G. 1985. Vegetation and Climates of the Last
45,000 Years in the Vicinity of the Nevada Test Site,
South-Central Nevada. U.S. Geological Survey, Reston,
VA, Professional Paper 1329, pp. 1-83.
. 1990. Comparison of pollen and macrofossil based
reconstructions of late Quaternary vegetation in west-
ern North America. Rev. Palaeobot. Palynol. 64:
359-366.
-, J. L. Betancourt, L. K. Croft, and K. L. Cole. 1990.
Packrat middens: their composition and methods of
analysis. In: J. L. Betancourt, T. R. Van Devender, and
P S. Martin (eds.). Packrat Middens. University of Ari-
zona Press, Tucson, AZ. pp. 59-84.
Stahle, D. W, M. K. Cleaveland, and J. G. Hehr. 1988. North
Carolina climate changes reconstructed from tree
rings: A.D. 372 to 1985. Science 240: 1517-1519.
Staplin, E L. 1962. Organic remains in meterorites—a re-
view of the problem. J. Alberta Soc. Petrol. Geol. 10:
575-580.
Steam, W. T. 1983. Botanical Latin. David and Charles Pub-
lishers, London.
Stockton, C. W., W. R. Boggess, and D. M. Meko. 1985. Cli-
mate and tree rings. In: A. D. Hecht (ed.), Paleoclimate
Analysis and Modeling. Wiley-Interscience, New York,
pp. 71-151.
Stone, D. E., S. C. Sellers, and W. J. Kress. 1979. Ontogeny
of exineless pollen in Heliconia, a banana relative.
Ann. Missouri Bot. Garden 66: 701-730.
. 1981. Ontogenetic and evolutionary implications
of a neotenous exine in Tapeinochilos (Zingiberales:
Costaceae) pollen. Am. J. Bot. 68: 49-63.
Stuiver, M. and R. L. Burk. 1985. Appendix: Paleoclimatic
studies using isotopes in trees. Appendix to Stockton
et al., Climate and tree rings. In: A. D. Hecht (ed.), Pa-
leoclimate Analysis and Modeling. Wiley-Interscience,
New York. pp. 151-161.
Swetnam, T. W. and J. L. Betancourt. 1990. Fire-southern
oscillation relations in the southwestern United
States. Science 249: 1017-1020.
Thomasson, J. R. 1983. Carexgraceii sp. N., Cyperocarpus
eliasii sp. N., Cyperocarpus terrestris sp. N., and Cype-
rocarpus pulcherrima Sp. N. (Cyperaceae) from the
Miocene of Nebraska. Am. J. Bot. 70: 435-449.
. 1984. Miocene grass (Gramineae: Arundinoideae)
leaves showing external micromorphological and in-
ternal anatomical features. Bot. Gazette 145: 204-209.
-, M. E. Nelson, and R. J. Zakrzewski. 1986. A fossil
grass (Gramineae: Chloridoideae) from the Miocene
with Kranz anatomy. Science 233: 876-878.
Thompson, R. S. 1985. Palynology and Neotoma middens.
Am. Assoc. Stratigr. Palynol. Contrib. Ser. 16: 89-112.
Todzia, C. A. and R. C. Keating. 1991. Leaf architecture of
the Chloranthaceae. Ann. Missouri Bot. Garden 78:
476-496.
Tomlinson, R B. and E C. Craighead, Sr. 1972. Growth-ring
studies on the native trees of sub-tropical Florida. In:
A. K. M. Ghouse and M. Yunus (eds.), Research Trends
in Plant Anatomy, K. A. Chowdhury Commemoration
Volume. Tata McGraw Hill, New Delhi, pp. 39-51.
Traverse, A. 1988. Paleopalynology. Unwin Hyman,
Boston.
, K. H. Clisby, and E Foreman. 1961. Pollen in
drilling-mud "thinners," a source of palynological con-
tamination. Micropaleontology 7: 375-377.
Upchurch, G. R., Jr. and J. A. Wolfe. 1987. Mid-Cretaceous
to Early Tertiary vegetation and climate: evidence from
fossil leaves and woods. In: E. M. Friis, W. G. Chaloner,
and P R. Crane (eds.), The Origins of Angiosperms and
Their Biological Consequences. Cambridge University
Press, Cambridge, U.K. pp. 75-105.
Urey, H. C. 1947. The thermodynamic properties of iso-
topic substances. J. Chem. Soc. Lond. 562-581.
Van Der Burgh, J., H. Visscher, D. L. Dilcher, and W. M.
Kiirschner. 1993. Paleoatmospheric signatures in Neo-
gene fossil leaves. Science 260:1788-1790.
Van der Water, R K., S. W. Leavitt, and J. L. Betancourt.
1994. Trends in stomatal density and 13C/12C ratios of
Pinus flexilis needles during last glacial-interglacial
cycle. Science 264: 239-243.
Van Devender, T. R. 1988. Pollen in packrat (Neotoma) mid-
dens: pollen transport and the relationship of pollen to
vegetation. Palynology 12: 221-229.
. 1990. Late Quaternary vegetation and climate of
the Chihuhuan Desert, United States and Mexico. In:
J. L. Betancourt, T. R. Van Devender, and R S. Martin
(eds.), Packrat Middens, the Last 40,000 Years of Biotic
Change. University of Arizona Press, Tucson, AZ. pp.
104-133.
et al. (+ 7 authors). 1985. Fossil packrat middens
and the tandem accelerator mass spectrometer. Nature
317:610-613.
Vaughan, T. A. 1990. Ecology of living packrats. In: J. L. Be-
tancourt, T. R. Van Devender, and P S. Martin (eds.),
Packrat Middens, the Last 40,000 Years of Biotic
Change. University of Arizona Press, Tucson, AZ. pp.
14-27.
von Post, L. 1916. Om skogstradpollen i sydsvenska torf-
mosslagerfoljder. Geol. Foren. Stockholm Forhdl. 38:
384-390. [See Davis and Faegri (1967) for English
translation.]
Webb, L. J. 1959. A physiognomic classification of Aus-
tralian rain forests. J. Ecol. 47: 551-570.
Wheeler, E. A. 1991. Paleocene dicotyledonous trees from
Big Bend National Park, Texas: variability in wood
types common in the Late Cretaceous and Early Ter-
tiary, and ecological inferences. Am. J. Bot. 78:
658-671.
and P. Baas. 1991. A survey of the fossil record for
dicotyledonous wood and its significance for evolu-
tionary and ecological wood anatomy. Internatl. Assoc.
Wood Anat. Bull. 12: 275-332.
-. 1993. The potentials and limitations of dicotyledo-
nous wood anatomy for climatic reconstructions. Pa-
leobiology 19: 487-498.
Wiles, G. C., P. E. Calkin, and G. C. Jacoby. 1996. Tree-ring
analysis and Quaternary geology: principles and re-
cent applications. Geomorphology 16: 259-272.
Wing, S. L. 1987. Eocene and Oligocene floras and vegeta-
tion of the Rocky Mountains. Ann. Missouri Bot. Gar-
den 74: 748-784.

. 1992. High-resolution leaf x-radiography in sys-
tematics and paleobotany. Am. J. Bot. 79: 1320-1324.
Wolfe, J. A. 1971. Tertiary climatic fluctuations and meth-
ods of analysis of Tertiary floras. Palaeogeogr., Palaeo-
climatol., Palaeoecol. 9: 27-57.
. 1977. Paleogene Floras from the Gulf of Alaska Re-
gion. U.S. Geological Survey, Reston, VA. Professional
Paper 997, pp. 1-108.
. 1978. A paleobotanical interpretation of Tertiary
climates in the northern hemisphere. Am. Sci. 66:
694-703.
. 1979. Temperature parameters of humid to mesic
forests of eastern Asia and relation of forests of other
regions of the northern hemisphere and Australasia.
U.S. Geological Survey, Reston, VA. Professional Paper
1106, pp. 1-37.
. 1981. Paleoclimatic significance of the Oligocene
and Neogene floras of the northwestern United States.
In: K. J. Niklas (ed.), Paleobotany, Paleoecology, and
Evolution. Praeger Publishers, New York. Vol. 2, pp.
79-101.
. 1987. An overview of the origins of the modern
vegetation and flora of the northern Rocky Mountains.
Ann. Missouri Bot. Garden 74: 785-803.
. 1989. Leaf-architectural analysis of the Hamameli-
didae. In: P. R. Crane and S. Blackmore (eds.), Evolu-
tion, Systematics, and Fossil History of the Hamameli-
dae, Vol. 1: Introduction and "Lower" Hamamelidae.
Clarendon Press, Oxford, U.K. Systematics Association
Special Vol. No. 40A, pp. 75-104.
. 1990. Estimates of Pliocene Precipitation and Tem-
perature Based on Multivariate Analysis of Leaf Phys-
iognomy. U.S. Geological Survey, Reston, VA. Open
file report 90-64, pp. 39-42.
. 1993. A method of obtaining climatic parameters
from leaf assemblages. U.S. Geol. Surv. Bull. 2040:1-71.
. 1994. Alaskan Palaeogene climates as inferred from
the CLAMP database. In: M. C. Boulter and H. C.
Fisher (eds.), Cenozoic Plants and Climates of the Arc-
tic. NATO ASI Series 1: Global Environmental Change.
Springer-Verlag, Berlin. Vol. 27, pp. 223-237.
-, A. B. Herman, and R. A. Spicer. 1996. Inferring cli-
matic parameters for Cretaceous and Tertiary leaf as-
semblages using CLAMP (climate-leaf analysis multi-
variate program). International Organization of
Palaeobotany Fifth Quadrennial Conference, Santa
Barbara, CA, June/July, 1996. p. 113.
Wolfe, J. A. and D. M. Hopkins. 1967. Climatic changes
recorded by Tertiary land floras in northwestern North
America. In: K. Hatai (ed.), Tertiary Correlations and
Climatic Changes in the Pacific. Sendai, Sasaki, Japan,
pp. 67-76.
Wolfe, J. A. and G. R. Upchurch, Jr. 1987. North American
non-marine climates and vegetation during the Late
Cretaceous. Palaeogeogr., PalaeoclimatoL, Palaeoecol.
61: 33-77.
General Readings
Chasan, R. 1996. What fossil woods say about climate. Bio-
Science 46: 803-804.
Keir, R. 1992. Packing away carbon isotopes. Nature 357:
445-446.
Methods, Principles, Strengths, and Limitations 141
Kern, R. A. and W. H. Schlesinger. 1992. Carbon stores in
vegetation. Nature 357: 447-448.
Lewin, R. 1985. Plant communities resist climatic change.
Science 228: 165-166 (discusses vegetational change
in the Grand Canyon region over the past 24,000 years
via analysis of packrat middens).
Niklas, K. J. 1987. Aerodynamics of wind pollination. Sci.
Am. 257: 90-95.
Visscher, H. 1997. LPP part two: paleophysiology. Am.
Assoc. Stratigr. Palynol. Newsletter 30: 9-10.
Yamaguchi, D. K., B. E Atwater, D. E. Bunker, B. E. Benson,
and M. S. Reid. 1997. Tree-ring dating the 1700 Casca-
dia earthquake. Nature 389: 922-923.
Notes
1. Claims of having recovered fossil organic microstruc-
tures from unanticipated sources (viz., metamorphic rocks;
meteorites; see references in Kremp, 1983; Staplin, 1962)
provide the lore and legend that is an entertaining part of
any field. In 1970 I received a rock sample from Professor
Wilhelm Klaus of the University of Vienna who asked to
have it processed for pollen and spores. It was clearly a
highly indurated metamorphic rock that proved virtually
impossible to mechanically or chemically disintegrate and,
as expected, it was barren of palynomorphs. Professor
Klaus was an eminent paleobotanist and certainly aware
that pollen and spores are not preserved in highly meta-
morphosed rocks. The explanation came when he in-
formed me later that palynologists working in one of the
former Eastern Block countries claimed to have developed
a processing technique whereby plant microfossils could
be obtained from such rocks. This claim reached the Direc-
tor at the Geologische Bundesanstalt who inquired why
scientists in other countries could do this while his staff
could not. Professor Klaus divided a sample into sections,
sent one to each of several laboratories throughout the
world, and requested a report be sent to the Director. All
the samples were barren and the matter was settled.
The novel technique was based on the assertion that pa-
lynomorphs were often present in metamorphic rocks but
were so delicate that they were destroyed by standard pro-
cessing procedures. The suggested alternative was to place
the rock in a beaker of water, set it on a window sill or, even
better, outdoors, and let the ultraviolet rays of the sun gen-
tly decompose the sample. The procedure worked best
when the beaker was uncovered. The palynomorphs were
remarkably well preserved and quite modern in aspect.
Years later on a congress cruise on the River Ob in Siberian
Russia, I talked with Professor Elena Zaklinskaya of the
USSR, who knew the details of the procedure. She indi-
cated that, as expected, the technicians were well aware of
the deficiencies and had used this obvious ruse to placate
overzealous supervisors then anxious for any claim of
technological superiority over the West.
2. Wolfe's (1993) classification of leaf-size categories is
leptophyll 1,2; microphyll 1, 2, 3; and mesophyll.
3. Taphonomy is the study of the transition of organic
remains from the living organism to fossil assemblages,
that is, events involved in the fossilization process (Efre-
mov, 1940; cited from Greenwood, 1991; see also Gastaldo,
1988).
 FIVE
Late Cretaceous through Early Eocene
North American Vegetational History
70-50 Ma
CONTEXT SUMMARY
At the end of the Cretaceous the Appalachian Mountains
had undergone 180 m.y. of erosion since their principal up-
lift in the Middle Pennsylvanian through the Late Permian
(300-250 Ma), but they were higher and more rugged than
at present and provided a somewhat more diverse verti-
cally zoned array of habitats (Figs. 5.1, 5.2). In contrast, the
Rocky Mountains were only -1 km above sea level at 65
Ma; the Coast Ranges, Sierra Nevada, and Cascade Moun-
tains would not attain substantial heights until late in the
Tertiary.
Computer models in the NM mode, which simulate
conditions in western North America in the Late Creta-
ceous, show a nearly continuous westerly jet stream with
relatively small amplitude between the troughs (low-
pressure systems) and ridges (high-pressure systems). The
present north-south seasonal meandering of the jet stream
was also less. Thus, in the models precipitation and tem-
peratures were more uniform throughout the year and
there was less regional differentiation in climate.
CO2 concentrations during the Cretaceous are estimated
to have been 4-8 times to 10-12 times higher than at pres-
ent. With a 2-5°C warming for each doubling of CO2, this
provides part of the explanation for the higher MAT docu-
mented for the Late Cretaceous and Early Tertiary.
High CO2 concentrations near the end of the Cretaceous
may have been a holdover from earlier intense volcanism
in the South Pacific that began to subside at ~ 100 Ma. The
term epeirogeny refers to vertical motions of the Earth's
crust, and these movements affect the ocean floor, as well
as the continents. There was a 50% increase in the produc-
tion of ocean crust in the Middle Cretaceous compared to
earlier times, as represented by the early Aptian Ontong
Java Plateau, the Earth's largest oceanic plateau, now sub-
merged over 2 km off the Solomon Islands. A sense of the
142
magnitude of this structure can be gained by comparing its
volume with that of the surface-exposed Deccan Traps of
India (66 Ma). The latter are -1 km thick and have a vol-
ume of 1.5 x 106 km3. The Ontong Java Plateau is -36 km
thick and has a volume of 50 x 106 km3. It is estimated to
have formed in just a few million years and was only part
of the midocean ridge, oceanic plateau, and flood basalt ac-
tivity of the Middle to Late Cretaceous. The speculated ori-
gin of the structure is that a part of the lower mantle broke
away and moved upward as a superplume beneath the Pa-
cific. The waning of volcanism after -100 Ma, and the as-
sociated reduced levels of CO2, contributed to the decline
of Late Cretaceous temperatures (Fig. 3.1). Although the
oxygen isotope paleotemperature curve reveals that tem-
peratures declined through the Late Cretaceous, values for
benthic ocean waters were still 10-12°C higher at the K-T
boundary than the present 1-2°C.
The term equable (warm, low seasonality) is often used
to describe Cretaceous climates, and there was only slug-
gish equator to pole atmosphere and ocean circulation
(Barron, 1983). Axelrod (1992a) notes the need for a better
definition of the term, and recent modeling results show it
may imply an overly simplistic view of Cretaceous and
Early Tertiary environments. GCM simulations demon-
strate that in the highlands at some far-northern interior
sites, continental configuration and size facilitated a mod-
erate ocean circulation and seasonal temperature regime,
possibly reaching near freezing (Barron and Washington
1982a,b; Barron et al., 1980, 1981; Crowley et al, 1986).
Rea et al. (1985) interpret eolian dust records to indicate an
atmospheric circulation as vigorous as in the past few mil-
lion years. Thus, the Hadley Cell was operating in the Late
Cretaceous even though the continents were arranged dif-
ferently, and there were extensive epicontinental seas (Hor-
rell, 1991; Zeigler et al., 1984). However, as noted by Sloan
and Barron (1990), models that predict Cretaceous and
 Late Cretaceous through Early Eocene History 143

Figure 5.1. Paleocene time scale. Reprinted from Berggren et al. (1995) with the permission of the Society for
Sedimentary Geology.

Early Tertiary temperatures remained substantially below
freezing for long periods of time over extensive areas, even
at the high interior northern latitudes, are not consistent
with a wealth of paleontological data (Herman and Spicer,
1996). It is true that many of these data come from warmer
coastal sites, as opposed to inland regions, but existing in-
formation is difficult to reconcile with frequent and pro-
longed periods of freezing. It has been suggested that pale-
ontologists reevaluate their data and interpretations of the
Cretaceous as essentially ice free (Barron et al., 1981). It is
also possible that the models may be depicting tempera-
tures that are too cold because coarse grid constraints do
144 Late Cretaceous and Cenozoic History of North American Vegetation

not fully incorporate local to regional physiographic fea-

tures (e.g., the paleo-Brooks Range). The Cretaceous plant
microfossil record from the North Slope of Alaska (Fred-
eriksen, 1989a; Frederiksen et al., 1988) includes over 100
different kinds of pollen and spores. This means that
forests and understory plants covered considerable parts of
the continental interior and that regional vegetation was
more extensive and diverse than suggested by some com-
paratively meager Arctic megafossil assemblages (Parrish
and Spicer, 1988). Associated faunal evidence (e.g., di-
nosaur remains from the North Slope) further document
the mesic nature of the climate. Computer simulations
show that forest cover theoretically can warm high-latitude
climates by 2.2°C, reducing discrepancies between the
observed and modeled latest Cretaceus climates (Otto-
Bliesner and Upchurch, 1997). The possibility, however, of
seasonal, cool-temperate environments in the interior of
the far north during the Cenomanian (Sellwood et al.,
1994) and Maastrichtian, and even local glacial conditions
above 1000 m at 85° N in the newly uplifted Brooks Range,
is not precluded by the paleontological data. Seasonality in
temperature, combined with a moderately diverse topogra-
phy, allows a greater array of habitats to accommodate the
more diverse vegetation now recognized from pollen evi-
dence than does an oversimplified generalization that cli-
mates were "equable" across the entire continent.
In the Late Cretaceous through at least Middle Eocene
times sea level was -300 m higher that at present, showing
a decline near the end of the Early Paleocene (60 Ma; Fig.
3.6). The highest levels of this interval would cover about
20% of the present Earth's surface. The high ocean level
was due primarily to Mesozoic plate reorganization and to
the absence of substantial glaciers. The formation of the
mid-Atlantic Ridge beginning in the Jurassic, and the uplift
of the Ontung Java Plateau in the Pacific, resulted in a de-
crease in ocean basin volume of about 2 x 1016 m3. These
events displaced water upon the continental margin and
contributed to the flooding of the low-lying interior region
of midcontinent North America. Global land-sea relation-
ships at 100 Ma are shown in Fig. 5.3. At various times dur-
ing the Cretaceous (e.g., in the Cenomanian), the waters of
the present Gulf of Mexico and the Arctic Ocean connected
to form a mostly shallow epicontinental sea that stretched
some 5000 km through the continent and partitioned North
America into an eastern and western portion. Among the
effects of this sea were milder climates, muted seasonal
variation in temperature, and the provincialization of the
North American flora into eastern and western provinces.
Reduced land surfaces and the extensive midcontinent
sea affected the Earth's albedo through greater absorption

Figure 5.2. Principal fossil floras mentioned in the text

plotted with general reference to the paleotemperature
record (lower curve) and sea-level record (upper curve) for
the Late Cretaceous through the Early Eocene. The Chalks
Bluff flora is now dated at ~48 Ma.
 Late Cretaceous through Early Eocene History 145

Figure 5.3. Generalized diagram of continental positions and land-sea relationships

(maximum depth of 100-200 m) at 100 Ma. Modified from Crowley and North (1991)
and based on Barren et al. (1980). Reprinted with permission of Elsevier Science-NL.

of solar heat. A sea level higher by 300 m would produce a
change in albedo of 0.03, which would translate into a rise
in surface MAT of ~3°C. More detailed calculations of the
Cretaceous albedo in the region of North America are spec-
ulative because the important variable of the extent and
thickness of cloud cover cannot be modeled accurately
(Tselioudis et al., 1993), and there are no paleoclimatic in-
dicators for cloud cover per se. If there was no cloud cover,
present ambient temperatures would be ~20°C higher be-
cause water vapor clouds (as opposed to CO2 clouds) are
more effective at reflecting solar radiation than trapping
heat through a greenhouse mechanism.
During the Mesozoic, North America was joined to Eu-
rope and Asia across the present North Atlantic and Pacific
Oceans. The Asian connection is evident, for example, from
the presence of Triceratops in eastern Asia and western
North America. By the Cretaceous the continents had
started to separate, and at -66 Ma they had the configura-
tion shown in Fig. 5.4. The North Atlantic connection had
only begun to fragment in the Maastrichtian as revealed by
the initial mixing of Arctic and Atlantic waters through the
Davis Strait. Thus, in terms of plant and animal migrations
there were no major physical barriers across the North At-
lantic. Climates likely included coastal subtropical zones
and warm-temperate to temperate ones locally in the north-
ern regions. Similar climates prevailed through Beringia.
Late Cretaceous and Paleogene faunas are known from
the margin of the epicontinental sea at localities from Texas
to the North Slope of Alaska. They are particularly abun-
dant and diverse from southern Kansas to southwestern
Saskatchewan. The first report of Mesozoic mammals, in
the form of abundant teeth, comes from the Lance Forma-
tion of Wyoming. Other faunas are known from the Hell
Creek Formation of eastern Montana and adjacent regions
and from the Scollard Formation of Alberta, Canada. The
latter includes the Bug Creek fauna, which is the richest
deposit of Mesozoic mammals in the world. These faunas
contain Allotheria (multituberculates), Metatheria (marsu-
pials), Eutheria (placentals), turtles, crocodilians, and
champsosaurs (Hutchison, 1982). The presence of non-
hibernating, nonburrowing species with limited migration
capabilities further implies warm climates without pro-
146 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 5.4. Plate reconstruction at chron 29 (Paleocene, 66.2 Ma). Reprinted from Scotese et al., (1989, in Scotese and
Sager, 1989) with the permission of Elsevier Science-NL.
longed periods below freezing. Coral reefs were present
5°-15° N of their present distribution (Habicht, 1979).
At the end of the Cretaceous an asteroid collided with
the Earth, leaving the Chicxulub crater now buried under
the Yucatan Peninsula. About 44% of the genera and 70%
of the species comprising the marine plankton and many
large terrestrial animals, notably the dinosaurs, became ex-
tinct at this time. In contrast, many other terrestrial groups
crossed the K-T boundary without extinction or their di-
versity had already begun to decline during the Late Cam-
panian and Maastrichtian (Frederiksen, 1989a). Paleocene
faunas still contain many elements found in the Late Cre-
taceous Bug Creek assemblage (Webb, 1989).
An asteroid impact is such an important and spectacu-
lar occurrence that it is tempting to interpret all kinds of
data as consistent with this highly publicized event. As the
duration of its effects are scaled down and its conse-
quences found to be selective, less tailored claims likely
will be forthcoming regarding the relationship between the
event and vegetational history. [For a balanced summary of
the spectrum of possible effects of the impact, see Crowley
and North (1991).]
The principal biological consequence of the terminal
Cretaceous event appears to have been extinction among
the groups noted previously. Its most intense climatic con-
sequences were brief (4-6 months), and the oxygen iso-
tope record does not detect any major ocean temperature
change bracketing the event. The impact may have tem-
porarily increased CO2 levels by 2-3 times (Hsu and
McKenzie, 1985) within an estimated concentration of
10-12 times that of modern values. The trend from ever-
green to deciduous vegetation resulting from a 4°C decline
in temperature between the Middle to Late Cretaceous (Fig.
3.1) may have been strengthened by an impact winter and
reduced light of 4-6 months' duration (Wolfe, 1987). The
effects are most evident in the midlatitudes of western
North America where the most complete fossiliferous tran-
sitional sections are available. The collision of an asteroid
with the earth at the end of the Cretaceous is convincingly
established. Present efforts are directed at assessing the rel-
ative roles of gradual environmental change and an abrupt
catastrophy on climate and biotas.
In the Paleocene through the Early Eocene, the Rocky
Mountains underwent their initial intensive folding to
about one-half or possibly locally to near their present av-
erage elevation. In the northern Rocky Mountains there
was considerable volcanic activity in the Early to Late
Eocene. Two effects of the development of these uplands
(to -2500 m or more) were to divide the previously contin-
uous lowlands of western North America into eastern and
western zones and to reduce the flow of moist Pacific
winds into the eastern Rocky Mountains and Plains area
(Wing, 1987).
North America remained connected to Europe and Asia
via the North Atlantic and Bering land bridges during the
Paleocene and Early Eocene. Even though Greenland was

Figure 5.5. Plate reconstruction at chron 25 (Eocene, 59.2 Ma). Reprinted from Scotese et al. (1989; in Scotese and Sager,
1989) with the permission of Elsevier Science-NL.
mostly separated from Europe (Fig. 5.5), the distance was
not great and migrations were facilitated by the mild cli-
mate. There were significant drops in sea level at 60, 49.5,
and 40 Ma, representing periods of relative quiescence in
ocean ridge activity and, at the latter dates, possibly early
glaciations on Antarctica. The epicontinental sea was re-
treating, and the High Plains were beginning to appear due
to continued uplift of the Rocky Mountains that tilted the
Plains area by ~1 km over a distance of -1400 km.
Drainage was also affected by continental rebound as the
sea disappeared and sediments began to erode. By the end
of the Early Eocene the inland sea had mostly disap-
peared, and Gulf and Atlantic waters occupied the areas
shown in Fig. 5.6.
Lower Tertiary faunas of southwest Texas include sharks,
skates and rays, tarpon, and crocodylidae; those from
Ellesmere Island included alligatorlike crocodilians, tur-
tles, and perissodactyl mammals. Collectively, the North
American Paleocene through Early Eocene faunas reflect
widely distributed tropical to warm-temperate elements
with strong affinities to Asia and Europe. As noted previ-
ously (Chapter 3), the Early Eocene represents the strongest
mammalian similarity that ever existed between North
America and Europe. There was little or no interchange as
yet with the faunas of South America.
Ocean temperatures increased and reached a maximum
for Cenozoic time in the Early Eocene (56 Ma; Fig. 3.1). The
values are estimated at between 11 and 15°C warmer than
Late Cretaceous through Early Eocene History 147
at present for benthic waters and 21°C for surface waters in
the high latitudes. There is no evidence for substantial con-
tinental glaciers during the interval between the Late Cre-
taceous and the end of the Early Eocene.
This varied and independent evidence provides a valu-
able set of parameters for interpreting the Late Cretaceous
through Early Eocene vegetation of North America. It
would be expected that the communities had broad geo-
graphic affinities, facilitated by the Thulean and DeGeer
routes to eastern Laurasia and the Beringian land connec-
tion to eastern Asia, although North America itself was
divided into eastern and western provinces by the Cre-
taceous midcontinental sea. The vegetation grew under
MATs that were significantly warmer than at present.
Under these conditions communities should be distributed
along low thermal gradients from south to north with
broad transitions and less differentiation into climatically
or altitudinally zoned communities, except in the Appa-
lachian Mountains that provided a moderate diversity of
altitudinally zoned habitats. As a rough approximation of
the south to north temperature cline, modeling results es-
timate a MAT of 23°C at 30° N latitude and 16°C at 50° N
latitude (Thompson and Barren, 1981). Isotopic evidence
yields a cline from 27 to 16°C, while present values at the
same latitudes are from 20 to 6°C. Lloyd (1982) estimates
the Cretaceous 18°C winter isotherm occurred at 40-50°
latitude from the equator, compared to its present position
of -30°. Paleobotanical data suggest a +5°-10°C difference
T 48 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 5.6. Approximate position of the Gulf-Atlantic
coastline at the end of the Early Eocene. Based on the Ge-
ologic Map of North America compiled by Marshall Kay.
at paleolatitude (pi) 30° N (Upchurch and Wolfe, 1987) and
+30°C at pi 80° N (Spicer and Parrish, 1986a), compared to
modern surface air temperatures.
VEGETATION CLASSIFICATION
Paleobotanical information necessary for tracing the his-
tory of North American vegetation is not uniformly avail-
able either stratigraphically or geographically. As a conse-
quence, this history must at times be extrapolated from
existing data elsewhere or inferred from the various con-
texts previously discussed. This is particularly true for the
Maastrichtian and Paleogene where a larger number of flo-
ras are known from the Western Interior basins than from
eastern North America.
Interpretation of the plant data is facilitated when bio-
logical affinities can be established for the fossils to at least
the level of family or genus. Artifical or form names must
be used for fossils beyond the range of extant taxa or where
the modern analog is not known. In many paleopalynolog-
ical studies the primary goals are age determination, bio-
stratigraphy, and correlation. For these geological applica-
tions, the biological affinities of the specimens are of
secondary importance (Leffingwell, 1971; Norton and Hall,
1967, 1969; Oltz, 1969; Snead, 1969; Srivastava, 1970;
Standley, 1965; Sweet, 1978; Tschudy, 1970). Palynological
studies as part of petroleum industry research are propri-
etary, and the published material is often limited to taxon-
omy and general discussions of stratigraphy and deposi-
tional environments. Thus, even though paleobotanical
information may be available for a particular age or local-
ity, it is not always in the form necessary for tracing vege-
tational or phylogenetic histories; this is especially true for
Late Cretaceous and Early Tertiary floras.
Another factor influencing the data base is the new
more inclusive approach to the study of many Late Creta-
ceous and Tertiary floras. At the time of E. W. Berry and
other early workers, it was conventional to consider fossil
floras mostly as separate units with the goal of expedi-
tiously providing a statement of the composition and pa-
leoenvironment for each assemblage (viz., paleofleristics).
One consequence of this approach was that variation within
the same taxon from distant sites was not immediately ev-
ident, which contributed to a proliferation of names and a
confused nomenclature. Also, modern methodologies such
as electron microscopy, X-ray radiograms, and cleared leaf
and pollen—spore reference collections were mostly not
available. Identifications were frequently based on margin-
ally preserved specimens, and many Cretaceous and Paleo-
gene plants were erroneously assigned modern family or
generic names on the basis of superficial resemblances in
morphology. Knowledge of the distribution and ecological
characteristics of the modern analogs, particularly for trop-
ical plants, the dynamics of vegetation change (e.g., the flu-
idity of community composition over time), plate tectonics
(for biogeographic studies essentially a post-1970 data
base), and comparative material from extant plants were all
meager by modern standards. The holdings at the United
States National Herbarium, used by E. W. Berry for his
study of southeastern United States and tropical American
Tertiary plants, numbered -1,223,400 specimens in 1925,
compared to the present ~4,500,000 accessions. For these
and other reasons the species lists for many of the older
megafossil floras are suspect.
A recent trend is to focus on lineages preserved across

several floras, ages, and geographic regions. Although the
initial aim may be the study of an individual flora, when
exceptional material is encountered it is often used as a
basis for investigating the worldwide fossil record of the
lineage with the goal of providing a statement of its origin,
phylogenetic relationships, and biogeographic history and
to provide a more sound nomenclature (viz., paleomono-
graphic studies). An example is the current project by
David Dilcher and coworkers on the Middle Eocene Clai-
borne and older associated ("Wilcox") floras of the south-
eastern United States. The floras are being revised, but pa-
leomonographic works are published as important material
is discovered. Among these are treatments of the Poaceae
(Gramineae; Crepet and Feldman, 1991), Gentianaceae
(Crepet and Daghlian, 1981), Juglandaceae (Manchester,
1987), and Malpighiaceae (Taylor and Crepet, 1987). These
paleomonographs are often done in collaboration with pa-
leobotanists in several parts of the world and involve taxo-
nomic specialists in the extant group (e.g., Quercus;
Crepet, 1989; Crepet and Nixon, 1989a,b; Nixon, 1989;
Nixon and Crepet, 1989). Similarly, Wolfe and Tanai (1987)
have integrated information from fossil and extant Acer in
a proposed new phylogeny and subgeneric classification,
and Patrick Herendeen is providing similar information for
the Leguminosale (see References, Chapter 6). A perusal of
recent papers by Ruth Stockey and her coworkers on the
Princeton flora of Alberta, Canada (Chapter 6), is witness to
the varied techniques, time, and meticulous care now de-
voted to the identification of fossil plant remains. Such
studies are of considerable value in providing a better un-
derstanding of the lineages and a reliable species list for
the floras. In the meantime it is important to recognize that
many North American Late Cretaceous and Tertiary floras
are presently a mixture of forms that have been studied re-
cently and intensely, together with those that have not
been revised for 75 years or more.
In describing the past vegetation and environments of
North America it is useful to follow a standardized termi-
nology as far as possible. Tropical, subtropical, temperate,
and cold temperate give a general but unquantified impres-
sion of climates; a comparable array of terms is applied to
moisture regimes and vegetation types. Greater consistency
in the way these terms are used is desirable and essential
for statistical treatment of the data. On the other hand, it is
futile to attempt overly precise definitions of entities and
events that are by nature fluid and transitional.
Systems of vegetation classification have importance be-
yond the naming of communities. If versions of essentially
modern vegetation types are viewed as extending back into
the Paleogene, names such as tropical rain forest, subtrop-
ical forest, deciduous forest, mixed conifer forest, and var-
ious combinations will be used. This terminology is com-
patible with the concept of geofloras moving over the
landscape in response to climatic change. If the paleo-
botanical record is interpreted as reflecting more fluid and
dynamic assemblages, names will be preferred that imply
Late Cretaceous through Early Eocene History 149
the temporal nature of paleocommunities. The Late Creta-
ceous vegetation in western North America between lati-
tudes 40°-50° N and 60°-65° N can be considered an eco-
tonal community between a subtropical evergreen and a
deciduous forest, or it can be called a notophyllous broad-
leaved evergreen forest. Concepts and semantics are often
interwoven, and the system of classification can be used to
imply a general similarity or a fundamental difference be-
tween paleovegetation and the modern flora. From the
standpoint of vegetational history it is convenient to pro-
ject modern vegetation formations as far back into the past
as is justifiable, as do Axelrod et al. (1991), because it keeps
a focus on the formation, its origin, and its development
through time. It is also true that the structure and composi-
tion of vegetation have changed over time, and at some
point the nomenclature should reflect these differences.
The transition is variably around the Early Eocene, as re-
vealed by the increasingly modern aspect of many
megafossils and palynomorphs at the generic level in Mid-
dle Eocene and later times. A special nomenclature for the
older communities has been developed by Wolfe, and it is
used here to convey their ancient and unique character.
Tropical forest—MAT ~25°C at the transition from trop-
ical to paratropical rain forest (Asia; Wolfe, 1979, p. 7),
subhumid, estimated rain fall < 1650 mm, little season-
ality, growth rings absent to poorly developed; an ex-
tinct, broad-leaved evergreen, single-tiered, open
canopy vegetation, leaves mostly entire-margined, thick
textured, few drip tips, leaf size indices below ~30.
Tropical rain forest—mean temperature of the coldest
month does not fall below ~18°C, MAT generally above
~25°C, no pronounced or extended dry season; broad-
leaved, evergreen, multistratal; drip tips, lianas, and
buttressing common; leaves sclerophyllous, mostly
mesophylls (megaphylls in substratum), entire-margined
leaves average over ~75%.
Paratropical rain forest (= near-tropical; subtropical of
many authors)—may experience some frost (-1°C to
-3°C), MAT ranges from 20°C to 25°C, precipitation may
be seasonal but there is no extended dry season; floristi-
cally allied to the tropical rain forest, predominantly
broad-leaved evergreen with some deciduous plants,
woody lianas diverse and abundant, buttressing pres-
ent; entire-margined leaves -57-75%.
A distinction frequently made between tropical and
paratropical forests is that the former is typically 3-
tiered while the latter has two canopy layers. This was
based on the classic work of Richards (1952), but a 3-
tiered canopy is no longer viewed as a defining feature
of the tropical rain forest. The two forest types are simi-
lar, particularly as represented in the fossil record.
Subtropical forest—frosts are present but not severe,
MAT is between 13 [and] 18°C, mean of the coldest
month is between 0 [and] to 18°C, rainfall is more sea-
sonal; sclerophylls are abundant, there are few lianas
and no buttressing, predominantly a broad-leaved ever-
150 Late Cretaceous and Cenozoic History of North American Vegetation

green forest with some conifers and broad-leaved decid- the Late Cretaceous of the Atlantic coastal plain and -19
uous elements; entire-margined leaves -39-55% (here from the Mississippi Embayment (Tschudy, 1975, 1980,
mostly included in paratropical rain forest or notophyl- 1981). Recently Areces-Mallea (1990) reported the Norma-
lous broad-leaved evergreen forest) polles Basopollis from the Paleogene of Cuba. The most
Notophyllous broad-leaved evergreen forest (ecotonal; widespread North American forms are Atlantopollis, Com-
oak-laural forest of eastern Asia)—mean of coldest plexiopollis, and Trudopollis (Fig. 5.9). Many Normapolles
month ~1°C, MAT ~13°C; some broad-leaved deciduous have smooth exines suggesting wind pollination. This
trees present, conifers not common, woody climbers mechanism is most typical of plants in open habitats with
abundant, buttressing virtually absent, sclerophyllous, at least moderate and/or seasonal dryness.
no drip tips, small mesophyll (notophyll) size class; In the west and extending into Asia the microfossil flo-
entire-margined leaves 40-60%. ras include a unique group of pollen types known as the
Warm temperate forest—temperatures fall below 0°C Triprojectacites. These are characterized by three promi-
during several months, there is a pronounced seasonal- nent equatorial and up to two polar projections (Fig. 5.10).
ity, mean annual temperature range is between 11 [and] They range in age from the Turonian (the oldest records are
13°C, mean of coldest month is between -3 [and] -2°C; from Siberia) through the Paleocene and reach their great-
a broad-leaved deciduous forest which may have a sig- est diversity in the Campanian. Farabee (1990, 1991, 1993)
nificant percentage of conifers, broad-leaved evergreens recognizes 14 morphotypes (approximate genera) among
are present but not dominant; entire-margined leaves 175 described species, and one genus (Aquilapollenites) is
-30-38%. used to designate the western North American-eastern
Polar broad-leaved deciduous forest—MAT ~7-8°C to Asian province. As a group, the Triprojectacites are poly-
15°C, distinct growth rings present; an extinct mesother- phyletic. The spinulose forms (e.g., Aquilapollenites) have
mal to microthermal moist forest type related to the some morphological similarities to the pollen of certain ex-
modern Deciduous Forest Formation; leaves large, thin- tant Loranthaceae and Santalaceae (Jarzen, 1977). In con-
textured. (Wolfe, 1977, pp. 24-26; 1979; 1981a, p. 82) trast to the Normapolles group, the Triprojectacites disap-
peared during the Early Tertiary. The southern terminus of
both provinces in North America is marked by tropical
Late Cretaceous Vegetation:
megathermal vegetation. In the far north the Tripro-
Maastrichtian, 70-65 Ma
jectacites are found within the primarily Normapolles
At the end of the Cretaceous (Fig. 5.7, Table 5.1) the North province (Greenland, the North Sea), suggesting that some
American continent was divided by an epicontinental sea parent plants may have been circumpolar.
(Jorgensen, 1993) into two floristic provinces (Batten, 1984; The Late Cretaceous and earliest Tertiary plant fossils in
Frederiksen, 1987; Herngreen and Chlonova, 1981; Srivas- North America were derived primarily from vegetation
tava, 1981, 1994a; Fig. 5.8). In the east and extending into growing along the coastal plains and slopes of the Ap-
western Europe, the Normapolles group is characteristic of palachian Mountains, Rocky Mountains, and the Alaska
Middle to Upper Cenomanian (Cretaceous) to Early Olig- and Brooks ranges. Their distribution essentially outlines
ocene deposits. Normapolles typically have three colpi or the margin of the epicontinental sea and adjacent oceans
pores that are structurally complex and frequently protrud- (Figs. 5.7, 5.8). In the southeastern United States the Mc-
ing (Traverse, 1988; Fig. 5.9). They represent a complex of Nairy Sand Formation (or Member of Berry's 1925 Ripley
early Hamamelidae (Betulaceae, Juglandaceae, the related Formation) of Tennessee (Tables 5.1, 5.2) is Maastrichtian
family Rhoipteleaceae, or an extinct family of the Juglan- in age; although it contains both leaves and wood, the ma-
dales; Friis, 1983). Near the end of the Cretaceous and into terial has not been studied taxonomically since Berry's
the Early Tertiary they became more diverse and widely (1925) original publication.1 Thus, it would be unwise to
distributed. In the Cenomanian worldwide there were attempt vegetational or paleoenvironmental reconstruc-
about six genera, while by the Maastrichtian there were tions on the basis of the fossils he assigned or compared to
more than 50. Of the latter, -26 have been reported from modern genera Acer, Aralia, Artocarpus, Bumelia, Cedrela,

Figure 5.7, facing page. Landform map of the conterminous United States with overlay of principal Late Cre-
taceous through Early Eocene floras and basins mentioned in the text (Thelin and Pike, 1991). (See also Figs.
2.13-2.14.) (1) US. Geological Survey's Oak Grove core (Paleocene), (2) Wilcox flora of western Tennessee (early
Eocene), (3) McNairy Sand Formation (Maastrichtian), (4) Rockdale Formation (microfossils; Paleocene), (5) Sen-
tinel Butte flora (Paleocene), (6-8) Golden Valley Formation (Bear Den and Camels Butte members; Late Paleocene
to Early Eocene), (9) Raton Basin (flora; Maastrichtian), (10) Denver Basin (flora; Maastrichtian), (11) Wind River
flora (late Early Eocene), (12) Willwood flora (Early Eocene), (13) Yellowstone floras (Middle Eocene), (14) Clark's
Fork flora (Paleocene-Eocene), (15) Fort Union floras (and eastward into the Dakotas; Paleocene), (16) Chalk Bluffs
flora (Early Eocene), (17) Chuckanut flora (Paleocene?). a, Vermejo Basin; b, Laramie Basin; c, Lance Basin
152 Late Cretaceous and Cenozoic History of North American Vegetation

Table 5.1. Geographic distribution of some Maastrichtian through early Eocene floras of North America.

Maastrichtian
Southeast McNairy Sand Wolfe and Upchurch (1987a), Berry (1925)
West Greenland (?)Lower Atanekerdluk Heer (1883)
Western interior
Texas, New Mexico Raton, Vermejo, Dawson, Arkose Knowlton (1917, 1930)
Colorado Denver, Laramie Knowlton (1922), McGookey et al. (1972)
Wyoming, Montana Hell Creek, Lance (Colgate), Medicine Bow Dorf (1942), Nichols et al. (1990)
Alberta Coalspur (Scollard) Bell (1949)
Lower Edmonton, St. Mary River Bell (1949)
Upper Edmonton, Whitemud Berry (1935)
Horseshoe Canyon Serbet and Stockey (1991)
Pacific Coast Patterson Page (1979, 1980, 1981)
Alaska Tongue River Frederiksen (1989a)
Prince Creek Frederiksen et al. (1988), Parrish and Spicer (1988)

Paleocene
West Greenland Upper Atanikerdluk Heer (1883), Koch (1963, 1964), Pedersen (1976)
North Greenland Thyra 0 Boyd (1990)
Western interior
North Dakota Sentinel Butte Crane et al. (1990)
Montana, Wyoming, Ft. Union Brown (1962)
North Dakota
Alberta Genesee Chandrasekharam (1974)
Saskatchewan Middle Ravenscrag, Berry (1935)
Upper Ravenscrag Jarzen (1982a,b)
Ravenscrag Mclver and Basinger (1993)
Paskapoo Bell (1949)
Alaska Chickaloon Wolfe (1966, 1972)

Early Eocene
Southeast Wilcox (part) See text
Western interior
Wyoming Wind River Leopold and MacGinitie (1972)
Yellowstone Wheeler et al. (1977, 1978), Wing (1987)
Pacific coast
California Chalk Bluffs MacGinitie (1941)
Alaska Kushtaka Wolfe (1977, 1985)
Adapted from Crane (1987, appendix) and Wolfe and Upchurch (1987a, table IV).

Celtis, Chrysophyllum, Cinnamomum, Dalbergia, Eugenia,
Euphorbiophyllum (Fig. 5.11), Fagus, Ficus, Grewiopsis,
Juglans, Laurus, Liriodendron, Myrica, Nectandra, Pota-
mogeton, Rhamnus, Salix, and Zizyphus. Rather, the meth-
odology developed by Jack Wolfe, described in the previ-
ous section, provides a more reliable basis for estimating
climates and the type of vegetation. The Perry Place and
Cooper Pit localities within the McNairy Sand Formation
contain 53 and 50 species, respectively. Values for entire-
margined leaves are 69 and 62% (and exceed 70% in other
associated floras), most are thick textured, there are few
species with drip tips, and the leaf-size indices are 28 and
29. These figures suggest a tropical forest-woodland
(entire-margined leaves 57-75%; Fig. 5.12) dominated by
broad-leaved evergreens and a megathermal temperature
regime (MAT > 20°C) reaching 25°C during warm intervals.
Little seasonality is inferred from the general absence or
poor development of growth rings in the fossil woods, al-
though anatomical features alone provide conflicting inter-
pretations of Cretaceous climates (Wheeler and Baas, 1991).
Low to moderate precipitation is indicated by the compar-
atively low leaf-size index. Extant megathermal vegetation
with abundant rainfall and no pronounced dry season has
a leaf-size index of -75. Maastrichtian woods from south-
ern Illinois have no growth rings (Wheeler et al., 1987), and
wood anatomy suggests many eastern North American
specimens were large trees typical of tropical angiosperms.
Some trunks are up to a meter in diameter.
The type of vegetation suggested by these indices is not
a typical multistratal, closed-canopy tropical rain forest,
and it has no exact counterpart in modern New World com-
munities. It is thus designated as a tropical forest (T in Fig.
5.8). A subhumid, open canopy where most members re-
ceive full sunlight is needed to explain the low leaf-size
 Figure 5.S. Generalized paleogeo-
graphic diagram of North America
during the Late Cretaceous with ap-
proximate positions of paleolatitudes,
principal formations-fossil floras
mentioned in the text, and tempera-
ture zones. (1) McNairy; (2) Lower
Atanekerdluk; (3) Vermejo; (4) Raton;
(5} Denver; (6) Laramine; (7) Patter-
son; (8) Hell Creek-Lance (Colgate;
Johnson, 1996); (9) Scollard, Horse-
shoe Canyon; (10) Coalspur; (11)
Prince Creek (Colville River site). D,
polar broadleaved deciduous forest;
N, notophyllous broadleaved ever-
green forest; P, paratropical forest; T,
tropical forest. Area of continental in-
undation represents maximum extent
during Late Cretaceous times.

Figure 5.9. Trudopollis variabilis, a normapolles pollen from the Campanian

Coffee Sand Formation, Tennessee (Tschudy, 1975, pi. 16,fig.15).

Figure 5.10. Integricorpus rigidus, a triprojectate pollen from the Campanian, Mon-
tana-Canada. Reprinted from Farabee (1990, fig. 16) with the permission of Elsevier
Science-NL.
Table 5.2. Paleolatitudes and physiognomic data on North American Maastrichtian floras.
Assemblage Formation pi- No. Sp. Entire (%) Leaf Index

Perry Place McNairy Sand 37 53 69 28

Cooper Pit McNairy Sand 37 50 62 29
L. Atanikerdluk3 Unnamed 55 56 77 12
Coalspur Brazeau 66 6 84
—
Hell Creek Hell Creek 56 21 62 53
Lance Lance 52 40 58 38
Medicine Bow Medicine Bow 50 42 67 55
Littleton Denver 48 50 71 57
Laramie Laramie 48 55 71 52
Lower Raton Raton 46 43 72 34
Vermejo Vermejo 46 63 71 34

pi, Paleolatitude. Adapted from Wolfe and Upchurch (1987a, tables V, VIII, IX).
a
Lower Paleocene fide Koch (1964).

Figure 5.11. Euphorbiophyllum cretaceum (Euphorbiaceae)

from the Late Cretaceous Ripley Formation, western Tennessee,
illustrating the generalized nature and poor preservation of many
specimens assigned to modern taxa in the older literature (Berry,
1925, pi. XII, fig. 6).

Figure 5.12. Generalized paleovegetation diagram for North

America during the Early Paleocene. T, tropical rain forest;
P, paratropical rain forest; D, broad-leaved deciduous forest.
Reprinted from Upchurch and Wolfe (1987) with the per-
mission of Cambridge University Press.

index; but regions supporting such extant megathermal
vegetation typically have a pronounced dry season, the
trees are shorter, and they grade into a semideciduous for-
est. Wolfe and Upchurch (1987a) provide the following es-
timates of precipitation for this unusual dry but not dis-
tinctly seasonal open-canopy woodland of southeastern
United States. The estimates are based on climatic condi-
tions and vegetation in New Caledonia and the vegetation
of some porous sandy soils in the wet tropics, which have
an abundance of conifers and many evergreen angio-
sperms: an annual rainfall of less than 1650 mm, a differ-
ence of less than 100 mm between the wet and drier sea-
son, and the driest month receiving at least 30 mm of
rainfall. Recall from Chapter 4 that the suitability of a mod-
ern southern hemisphere analog for these Cretaceous floras
is unsettled. The Ripley flora has recently been compared
physiognomically to extant Sinaloan (Mexico) deciduous
forest (Wolfe, in press). The distribution of coals and evap-
orites (Frakes, 1979) reflects rainfall values that were low
in many regions during the Cretaceous and increased sig-
nificantly in the Paleocene. Other sedimentological pat-
terns are not so clear (e.g., Hallam, 1984), and interpreta-
tions from this line of evidence give a mixed picture of
rainfall regimes (Parrish, 1987). Crepet (1981) and Batten
(1984) also believe that the Late Cretaceous climate in the
southeastern United States was dry; but based in part on
the abundance of the presumably wind-pollinated Norma-
polles group, they also believe it was more seasonal (con-
sistent with recent CLAMP analysis; J. A. Wolfe, personal
communication, 1996). The early radiation of angiosperms
from west Gondwana occurred from similar open areas in
equable uplands with some seasonal drought (Axelrod,
1952, 1970; Stebbins, 1974). Wing and Tiffney (1987) cite
grazing by dinosaurs as a factor in keeping the vegetation a
low, serai forest, which they believe is indicated by small
diaspore size. Wolfe and Upchurch (1987a) believe this
small size is due to a dry climate. These differences in
interpretation emphasize the approximate nature of the re-
constructions. However, it is generally agreed that condi-
tions in the southeastern United States during the Maas-
trichtian were warm and dry with at least moderate
seasonality in rainfall.
To the far northeast there is a poorly known flora possibly
of Maastrichtian age from western Greenland. It is from an
unnamed formation and the depositional setting is not
known, although it may have been along a floodplain. The
Lower Atanikerdluk flora (Tables 5.1, 5.2) consists of 56
species; 77% have entire-margined leaves, requiring a
megathermal climate, but with a leaf-size index of 12. This
would indicate a tropical forest by Wolfe and Upchurch's
(I987a) definition, but the leaf-size index is anomalously
low for that community. Leaf fossils have been referred to
the form genera Proteoides, Dermatophyllites, and Chon-
drophyllum, and to the extant genera Magnolia, Androm-
eda, and Diospyros. All of these also occur in Late Creta-
ceous floras in the southeastern United States and reflect the
Late Cretaceous through Early Eocene History 155
continuity of vegetation growing along low thermal gradi-
ents into the high northern latitudes. Coastal environments
were certainly mild; but climatic reconstructions based on
the flora are speculative, and nothing is known about Maas-
trichtian vegetation from the interior of Greenland.
Woods of Maastrichtian age are known from Big Bend
National Park (Javelinoxylon, malvalean affinities; Wheeler
et al., 1994). Sedimentological evidence and isotopic
analysis of carbonate nodules from the paleosols suggest a
warm dry climate with a MAT exceeding 15°C (Ferguson et
al., 1991; Lehman, 1990).
On the western side of the epicontinetal sea, in the
Aquilapollenites province, remnants of the Late Cretaceous
and Early Tertiary vegetation are preserved in a series of
basins extending from the southwestern United States to
the North Slope of Alaska (Figs. 5.7, 5.8, Tables 5.1, 5.2).
Leaf assemblages from the Vermejo, Lower Raton (Up-
church, 1995), Laramie, and Lower Denver formations
were situated at about pi 40°-46° N in the Maastrichtian
(presently ~37°-43° N). The number of species, percentage
of entire-margined leaves, and leaf-size indices are given in
Table 5.2. The MAT is estimated at 21-22°C (low mega-
thermal) to 18°C (Nichols et al., 1989) for the Powder River
Basin of Wyoming. Moisture conditions were generally
subhumid (estimated -1200 mm, Powder River Basin), but
climates varied locally and/or over short spans of time. For
example, leaf-size indices are higher in the Denver flora
than in the Raton flora (800-1000 mm). The Denver assem-
blage also includes several species with moderate drip tips
and others represent lianas (Menispermaceae), suggesting
higher and less seasonal temperature and rainfall at this lo-
cality. Along the Pacific coast, woods of Maastrichtian age
are known from near Patterson in central California (pi 45°
N; Page 1979, 1980, 1981). Nine of the taxa exhibit no
growth rings while in one they are well developed. This
limited data from the coast suggests little temperature sea-
sonality, although some slight seasonal differences in rain-
fall may have occurred. In general, rainfall values appear to
be somewhat higher for western North America than in the
east, and winds off the Pacific Ocean may have been the
source. This paratropical megathermal vegetation (Fig. 5.8)
probably extended somewhat further north, perhaps to pi
60°-70° N along the coast where January mean tempera-
tures were likely above freezing.
It has long been recognized that a transition zone was
present between ~pl 40° N and 50° N (Dorf, 1942), repre-
senting the broad boundary between megathermal and
mesothermal temperatures (the 20°C isotherm is placed
within this zone; Fig. 5.8). The floras on either side are dif-
ferent in composition, and those to the north have better
representation of deciduous dicotyledons among the still
dominant evergreens. In general, woods at and below ~pl
45° N have no or poorly developed growth rings, while
those above ~pl 55° N mostly represent smaller trees and
large shrubs that have better developed growth rings par-
tially resulting from greater but slightly seasonal rainfall.
156 Late Cretaceous and Cenozoic History of North American Vegetation
Between pi 50° N and 58° N (viz., in the northern Rocky
Mountains) the vegetation was a notophyllous broad-
leaved evergreen forest growing under a mesothermal and
somewhat more moist but still subhumid climate than to
the south and southeast (Fig. 5.8). Lianas and leaves with
drip tips are rare and leaf size is relatively small, suggest-
ing only limited closed-canopy forests. Some Araucari-
aceae and evergreen Taxodiaceae were present, broad-
leaved deciduous vegetation grew mostly along streams,
and Cycadophyte and Ginkgophyte holdovers from the
Middle Cretaceous are better represented than in floras of
similar age to the south and southeast.
Leaves from Maastrichtian floras located between pi 58°
N and 66° N in the Western Interior have been described by
Bell [1949; Coalspur (Scollard), Lower Edmonton, St. Mary
River formations] and Berry (1935; Upper Edmonton, White-
mud formations). The assemblages include heterosporus
ferns (Hydropteris pinnata; Rothwell and Stockey, 1994),
aquatic angiosperms (Triapago angulata; Stockey and Roth-
well, 1997), broad-leaved deciduous angiosperms, a variety
of deciduous gymnosperms (Glyptostrobus, Metasequoia),
and a few broad-leaved evergreens (palms). The percentage
of lobed-leaved species is low, the leaves are generally of
thin texture and moderately large in size, and diversity is
low compared to modern extant microthermal broad-leaved
deciduous vegetation. These are features of an extinct veg-
etation type described as a polar broad-leaved deciduous
forest (Wolfe, 1985; Fig. 5.8) that grew under mesothermal
to microthermal, moist climates. The MAT is estimated at
~15°C (Wolfe and Upchurch, 1987a). The climate was a bal-
ance between the Middle Cretaceous thermal maximum,
declining temperatures of the Maastrichtian, and some
northward movement of the craton that included Alaska in
a more polar position.
Overall, fossil woods from the Maastrichtian of the
Western Interior are poorly known, and the inadquate
number of specimens studied gives a mixed climatic sig-
nal. Two conifer woods are known from the Vermejo For-
mation; in one growth rings are present and in the other
they are poorly developed to absent. Coniferous wood from
the Hell Creek Formation of eastern Montana has growth
rings.2 Ramanujan and Stewart (1969) describe four conif-
erous woods from the Maastrichtian of central Alberta [pi
65° N; Coalspur (Scollard)], all with well-developed
growth rings. These data are consistent with evidence from
Lance-Hell Creek mammals (the "Triceratops commu-
nity" of Van Valen and Sloan, 1977) that suggests condi-
tions warmer than those of the Bug Creek ("Protungulatum
community") assemblage from Alberta, Canada.
The boundary between mesothermal and microthermal
temperatures (MAT of 13°C) is placed at pi 65°-75° N. In
the far northwest, floras of Late Cretaceous to Paleocene
age are known from the North Slope of Alaska at pi
75°-85° N (the present Arctic Circle is at 66° 33' N). These
are the highest paleolatitude Maastrichtian floras known
for North America. The MAT is estimated at 10°C for the
Cenomanian, 13°C in the Coniacian, 2-8°C (average 5°C,
similar to the present-day MAT at Anchorage) in the Maas-
trichtian (~8°C by extrapolation of the latitudinal tempera-
ture gradient in Wolfe and Upchurch, 1987a), and 6-7°C in
the Paleocene (Parrish and Spicer, 1988; Spicer and Par-
rish, 1986a, 1990). The Coniacian MAT for the North Slope
of Alaska is now placed at ~12.5°C and the cold month
mean at 5.7°C. For air temperature to be maintained near
5.7°C during the dark winter months, SSTs are estimated at
between 6 and 8°C. The mechanism proposed for sustain-
ing these warm polar temperatures is a significant pole-
ward transport of heat (Herman and Spicer, 1996, 1997). If
the deciduous Taxodiaceae had about the same ecological
requirements as their modern descendants, the earlier
MAT estimate of 5°C for the Maastrichtian would be mini-
mum to slightly low because their vegetative and repro-
ductive buds would have been damaged by frosts that
would have occurred in the coldest month within a MAT
of 5°C. At the present ecotone between mixed deciduous
angiosperm and evergreen conifer forests at Prince Rupert,
British Columbia, the MAT is 7.6°C and the coldest month
average is 1.8°C (Horrell, 1991). Conditions were generally
cooler in the high latitudes of the west than in the east. The
precipitation trend was from driest in the southern United
States to wettest in the Arctic; a sharper increase in precip-
itation at pi 46°-48° N was reflected by larger leaf sizes.
Pollen of Cicatricosisporites, Sequoiapollenites, Taxo-
diapollenites, Aquilapollenites, and Cranwellia has been
reported from Late Cretaceous sediments in the Yukon-
Tanana region of Alaska (~64° N; Foster and Igarashi, 1990).
Frederiksen (1989a; see also Frederiksen et al., 1988) de-
scribed Campanian to Maastrichtian palynofloras from the
Tongue River Formation along the Arctic Slope of Alaska.
The megafossil record has been interpreted as a relatively
low-diversity angiosperm understory and a low-diversity
ground cover of ferns and Equisetum (Parrish and Spicer,
1988). The assemblage upon which this reconstruction was
based, however, consisted of only 10 taxa: two conifers,
three angiosperms, one sphenophyte, two ferns, and two
seed plants of unknown affinities (plus subsequently six
taxa of coniferous wood). The microfossil record includes
110 types from 133 pollen-bearing samples through a
stratigraphic section 1400 m thick. The value of the micro-
fossil assemblage is that it includes a broader sampling of
the regional vegetation and provides a more extensive
record of the regional forest (Betulaceae-Myricaceae, Ul-
maceae) and probable understory plants that are rare or ab-
sent as megafossils. The combined observations again sup-
port a synergistic approach for interpreting vegetational
and paleoenvironmental history. The greater diversity re-
vealed by the microfossils also argues for a MAT near the
higher range of the 2-8°C estimate. Clemens and Allison
(1985) reported remains of large dinosaurs from the Arctic
Slope, which probably could not have existed there if the
coldest month temperature was below ~7-8°C. However,
this gets into the debate as to whether these dinosaurs were
 Late Cretaceous through Early Eocene History 157

Table 5.3. Dicotyledons from Late Cretaceous/Maastrichtian Floras of North America.

Taxon Affinities Oldest Occurrence Reference
"Myrtophyllum" torreyi Magnoliidae LateK Brown (1962)
Magnoliales Upchurch and Dilcher (1990)
Fagopsis Hamamelidae, Late Campanian/early Wolfe and Upchurch (1986)
Fagales Maastrichtian
Aff. Fagaceae
"Carya" Hamamelidae Maastrichtian Hickey (1980)
antiquorum Juglandales
Juglandaceae
Aff. Averrhoites Rosidae Maastrichtian Wolfe and Upchurch (1987a)
related to Sapindopsis'?
"Acer" Rosidae Late Campanian or Early Wolfe and Upchurch (1986)
arcticum Sapindales, Maastrichtian
complex aff. Aceraceae Wolfe and Tanai (1987)
Theaceae Dilleniidae Campanian to B. H. Tiffhey (unpublished data)
Theales Maastrichtian J. A. Wolfe and G. R. Upchurch, Jr.
Theaceae (unpublished data)
"Cissites" Dilleniidae Maastrichtian Wolfe and Upchurch (1987a)
panduratus, Euphorbiales
"Ficus" leei Euphorbiaceae
Parabombacaceaeoxylon Dilleniidae Maastrichtian Wheeler et al. (1987)
Malvales no extant family
Tiliaceae Dilleniidae Maastrichtian Wolfe and Upchurch (1987a)
Malvales
Tiliaceae

Adapted from Upchurch and Wolfe (1993; see also Wolfe and Upchurch, 1986). For a listing of Maastrichtian plants from North America and other
areas, including palynomorphs, see Horrell (1991).

warm or cold blooded (Barrick and Showers, 1994) and if
they could have migrated seasonally (Axelrod, 1984). If
they did not, it is unclear how they spent their time in the
dark winter months (Crowley and North, 1991; but see
Clemens and Nelms, 1993). Older Late Cenomanian woods
show distinct growth rings with very narrow summer
wood, suggesting rapid onset of winter conditions (Spicer
and Parrish, 1986b). All the Late Cretaceous terrestrial veg-
etation of the North Slope of Alaska presently known was
deciduous, died back to underground organs, or survived
the winter as seeds (Spicer and Parrish, 1990), which sug-
gests a seasonal temperature—light regime within a cool-
temperate climate.
In the introductory section of this chapter it was sug-
gested on the basis of context information that North
American Late Cretaceous vegetation was distributed over
lower thermal gradients than at present. It is now possible
to quantify this gradient to some extent. Figure 5.8 shows
an estimated MAT of 23°C at pi 30° N and 8°C at pi 75° N.
Thus, the temperature values changed by ~15°C over -45°
of latitude for a lapse rate of ~0.3°C/1° latitude. This com-
pares to the steeper modern gradient of ~0.5°C/1° of lati-
tude (0.8-1.0° in eastern and central North America).
Other evidence also suggests the Late Mesozoic thermal
gradient was about one-half that of the present (Berggren,
1982). As noted previously, the Late Cretaceous boundary
between megathermal and mesothermal temperatures
(MAT 20°C) is placed between pi 40° N and 50° N, and be-
tween mesothermal and microthermal (MAT 13°C) it is
placed between pi 65° N and 75° N. This compares with
the present positions along the western coast of 30° N and
40° N, respectively (Upchurch, 1989).
The plants comprising North American vegetation dur-
ing the Maastrichtian (Table 5.3) are difficult to relate to
modern taxa, and the plant communities are difficult to re-
construct in detail because familiar modern analogs mostly
do not exist. The angiosperms certainly diversified in the
Late Cretaceous as shown by a doubling of extant families
represented in the pollen record compared to the Campan-
ian. All subclasses of the dicotyledons except the Asteridae
are present by the end of the Maastrichtian (Crane, 1987,
fig. 5.2d). Megathermal vegetation below pi 40°-50° N (Fig.
5.8) in both the east and west included a diverse gym-
nosperm element of Taxodiaceae [evergreens similar to Se-
quoia; deciduous forms assignable to Parataxodium (ex-
tinct) and Metasequoia] and possibly the Cupressaceae
(remains similar to the extant Callitris and Neocallitropis,
which presently has one species in New Caledonia). These
coniferous trees probably formed an emergent, open-
canopy woodland. Monocots were represented by palms,
probably in a rosette growth form, which first appeared in
the Santonian of megathermal regions and in the Campan-
ian of mesothermal regions, and by the Zingiberales (first
appearing in the Maastrichtian; Spirematospermum, Friis,
1987; Zingiberopsis, Hickey and Peterson, 1978). The di-
cots included members of the Magnoliidae, Hamamelidae
[Plantanaceae, Fagaceae (Fagopsis), Juglandaceae], Rosidae
(Aceraceae as represented by "Acer" arcticum, the inferred
158 Late Cretaceous and Cenozoic History of North American Vegetation
sister group of extant maples), Ranunculidae, and Dilleni-
idae. These formed a low, open understory beneath the
conifers through the Early Maastrichtian (Upchurch and
Wolfe, 1993). Vines were few, reflecting the subhumid con-
ditions. The older Cycadophytes and Ginkgophytes were
virtually absent from the southern region. By the Late
Maastrichtian, angiosperms were dominant in megather-
mal climates and may have been the principal canopy-
forming group in late successional to climax communities
throughout most of North America. However, Tiffney
(1984) and Crane (1987) believe the angiosperms were
mostly small plants during most of the Cretaceous and that
gymnosperms still played a prominent role. In the north-
ern transition zone at ~pl 40°-50° N, megathermal vegeta-
tion extended into regions of higher precipitation and
probably formed a more closed-canopy forest, as suggested
by the larger leaf size (~50 vs. 28-40). In assemblages
where facies analyses have been made, angiosperms were
dominant in many floodplain, swamp, and lacustrine habi-
tats by the beginning of the Late Maastrichtian.
Between pi 40°-50° N and 60°-65° N the vegetation
was a notophyllous broad-leaved evergreen forest. In the
Aquilapollenites province Araucariaceae and pollen of Tri-
colpites reticulatus, considered similar to Gunnera (Jarzen,
1980; Jarzen and Dettmann, 1989), have been reported. Per-
mineralized cones with pollen have been recovered from
the Maastrichtian Horseshoe Canyon Formation near
Drumheller, Alberta, Canada (Serbet and Stockey, 1991).
The cones are described as Drumhellera kurmanniae of the
Taxodium-Metasequoia-Sequoia-Sequoiadendron com-
plex that was abundant in swampy, wetland habitats. Other
remains include Cycadophytes, Ginkgo, Equisetum (horse-
tail), lycopod megaspores, and cercidiphyllaceous fruits
and leaves. Broad-leaved deciduous plants were common
in successional or disturbed habitats, especially along
streams. The presence of a few vines and some species
with drip tips and the somewhat larger size of the leaves
suggest slightly more humid conditions, and woods reflect
some seasonality in temperature or precipitation. In sev-
eral respects the notophyllous broad-leaved evergreen for-
est is similar to an ecotonal paratropical (subtropical) com-
munity near the limits of its northern distribution and one
transitional between subtropical forests to the south and
deciduous forests to the north.
North of pi 60°-65° N to as far as 80°-82° N, there was
a polar broad-leaved deciduous forest. Deciduous gym-
nosperms were common; angiosperms with large, thin-
textured leaves formed an understory vegetation indicative
of a microthermal climate and polar light regimes. The
presence of fusain (charred organic material) suggest that
fires may have contributed to an open-canopy structure
(Spicer and Parrish, 1990; Spicer et al., 1994). The emerg-
ing picture of Late Cretaceous climate at these high lati-
tudes is one that was somewhat cooler and more winter
seasonal than previously suggested (Smiley, 1967). This
coolness and its associated increase in wetness also better
accounts for the presence of extensive Late Cretaceous and
Paleocene peat and coal deposits on the North Slope. At
the same time it limits the role of frequent fires as an expla-
nation for the openness of the vegetation.
Plants of the polar broad-leaved deciduous forest were
dominated by angiosperms (since earlier Cretaceous times)
and included the Taxodiaceae as the most abundant and
widespread gymnosperm. As in the region immediately to
the south, Ginkgo was present in the southern part of the
polar deciduous forest, along with trochodendroids,
"Viburnum"-type leaves, platanoids (Herman, 1994), and
other Hamamelidae. Megafossils from the Kogosukruk
Tongue Member of the Prince Creek Formation along the
Colville River, North Slope of Alaska (Campanian and
Maastrichtian), also reveal a vegetation that included de-
ciduous coniferous trees. The microfossils add members of
the Betulaceae-Myricaceae and Ulmaceae complexes that
diversified during the Campanian, Maastrichtian, and
Paleocene, even though overall diversity in the far north
declined toward the end of the Maastrichtian. The decline
was due primarily to fewer pollen types of entomophilous
species (Frederiksen, 1991a), suggesting that temperatures
were near the minimum for pollinators.
Two problems that complicated reconstruction of far-
northern biotas earlier were recently resolved. A signifi-
cant alteration was proposed in the above-described south
(tropical) to north (temperate) pattern of North American
Late Cretaceous and Paleogene communities based on pa-
leomagnetic data from Campanian to Early Eocene de-
posits in Arctic Canada (Axel Heiberg, Ellesmere islands,
Bay Fiord area; Hickey et al., 1983). Fossils associated with
the Eureka Sound Formation were interpreted as docu-
menting latest Cretaceous megathermal plants in the Arctic
that preceded their midlatitude occurrences by -18 m.y.,
while Eocene vertebrate fossils supposedly appeared in the
Arctic 2-4 m.y. earlier than in the south. If so, the original
dispersal center for many tropical species would have to be
placed in the Cretaceous Arctic because of their earliest oc-
currence there. Studies have now shown that the parent
rocks containing the fossils had been altered by secondary
magnetization (magnetic overprinting), the stratigraphic
ranges used for certain palynomorphs were incomplete,
and some microfossils had been redeposited (Kent et al.,
1984; Norris and Miall, 1984; Opdyke, 1990). These plant-
bearing sections within the Eureka Sound Formation are
now regarded as Eocene in age and are compatible with the
generally accepted latitudinal distribution pattern of North
American paleocommunities.
A second problem was envisioning a forested vegetation
growing under the low light intensities and long periods of
darkness in the polar environment. At the high northern
latitudes above the Arctic Circle the sun is at to just below
the horizon for about 6 months of the year and at to just
above for the other 6 months (to a maximum of 43° during
the Arctic summer). The present vegetation is mostly tun-
dra. A higher CO2 concentration in the Cretaceous and Pa-

leogene has been proposed as one mechanism for in-
creased warmth and, as a consequence, for more available
moisture. The dim light, however, remained an enigma and
required some innovative thinking. An early suggestion
was that the tilt of the Earth's polar axis (presently at 23.5°)
may have been significantly less in the past (5°—15°) and
that it varied in concert with vegetational changes read
from the fossil record: "These are precisely the changes
that are inferred from paleobotanical data for the Oligocene
and Neogene and would indicate that a significant de-
crease in the earth's inclination has occurred during the
last 30 million years" (Wolfe, 1978, p. 701; 1980). However,
a significantly reduced obliquity has no known basis in ce-
lestial mechanics because a plausible mechanism for the
change is unknown, and modeling experiments actually
produced a substantial cooling (by ~10°C at an obliquity of
15°; Barren, 1984). Creber and Chaloner (1985) cite evi-
dence that the proposed analogy to obliquity variations on
Mars does not apply to Earth because of its different shape
and the stablizing influence of the moon. Other difficulties
have been reviewed by Axelrod (1984) and Crowley and
North (1991). The idea persists, however, and recently
Muller and MacDonald (1997) suggested that changes in
inclination periodically bring the earth into the cosmic
dust cloud which then alters climate. The study was pro-
posed as an alternative to or an amplification of the 100 Ky
Milankovitch eccentricity cycles of the Quaternary, but the
mechanism could apply to more ancient times. The theory
is still not widely accepted, but no alternative explanation
was immediately apparent to explain the existence of
forests under polar light regimes. An obstacle has always
been the underlying assumption that even with warmer
temperatures and more available moisture, present light
levels are too low to support trees. That assumption was
never tested, and the review by Creber and Chaloner (1985)
indicates that it may not be true. Accumulating evidence
from plant physiology suggests that the primary limitation
to large plants in the Arctic is low temperature, which re-
duces the rate of photosynthesis (see references in Creber
and Chaloner, 1985). Measurements of light levels show
that the June solar input on Cornwallis Island (27.8
MJ/m2/day; 75.15° N) is about the same as for Washington,
D.C. (31.5 MJ/m2/day; 38.54° N) because during the sum-
mer season at the poles there is sunlight over the full 24 h.
The calculations indicate that trees most suitable for inter-
cepting maximum amounts of limited light would be coni-
cal in shape and widely spaced. For a tree 17m tall with a
basal radius of 1.6 m, the crown would have a light-
intercepting surface of -25 m2. This surface would receive
over 80,000 MJ of radiation during the illuminated part of
the year under ideal conditions of no shading and limited
cloud cover, which is more than sufficient to support tree
growth. Measurements of tree density in Eocene floras of
Ellesmere Island (Iceberg Bay Formation) actually suggest
spacing about the same as in mesic forests of the midlati-
tudes (Francis, 1988). Nonetheless, the consensus is that a
Late Cretaceous through Early Eocene History 159
deciduous vegetation of trees and shrubs could have ex-
isted in the dim Arctic light under present inclination val-
ues, assuming higher temperatures and available moisture.
One of the major changes in the modernization of the
North American vegetation has been the replacement of
this Late Cretaceous and earliest Tertiary polar broad-
leaved deciduous forest and associated deciduous conifers
by the boreal coniferous forest of evergreen conifers (Abies,
Picea) in later Tertiary and modern times. As will be seen
later, this transition is evident at midaltitudes (700-900 m)
in the Middle Eocene of the Pacific Northwest (Thunder
Mountain flora of Idaho; Republic flora of Washington). It
intensified in the Late Eocene, consolidated during the
Oligocene, and was accomplished by the Mio-Pliocene.
THE K-T BOUNDARY: 65 MA
The bolide collision with the Earth at the end of the Creta-
ceous produced the terrestrial biological consequences dis-
cussed earlier (notably selective extinctions). Beyond this
clear effect, it is difficult to separate changes in terrestrial
vegetation at the K-T boundary due to the impact from
those resulting from erogenic and climatic change. Further
complications are that complete sections through the tran-
sition are not widely available, the uniformly drastic and
worldwide effects implied by the original scenario have
been scaled down, the effects differed at various sites and
latitudes, and broad-leaved evergreen megathermal vegeta-
tion suffered more than microthermal deciduous commu-
nities. The present data allow a range of interpretations re-
garding the immediate effect of the impact, as well as its
long-term consequences on climate and biotic history.
The marine 18O record is equivocal at the K-T bound-
ary. Isotopic studies show gradually cooling temperatures
across the boundary (Crowley and North, 1991; Savin,
1977). Thus, it is possible to read this record as a trend of
lowering temperatures that, along with the worldwide ma-
rine regressions and lowering CO2 levels, reached a thresh-
old that was primary in determining the vegetational his-
tory (Rickey, 1980,1981). Wolfe and Upchurch note that
[D]uring the Paleocene, a marked increase in diversity
of deciduous dicotyledons occurred in mesothermal
and adjacent megathermal vegetation. This anomalous
deciduousness resulted in vegetation that was not phys-
iognomically synchronized with temperature. Thus,
leaf-margin analyses of particularly mesothermal, and
to some extent megathermal, Paleocene vegetation yield
temperature estimates that are far too low, which have
resulted in invalid suggestions of a major temperature
decline from the late Maastrichtian into the Paleocene.
(I987a)
[See also Wolfe and Upchurch (1986).] If the impact is con-
sidered primary, then the 18O record can be interpreted as
reflecting a temperature lowering insufficient and too grad-
ual to produce the changes observed.
160 Late Cretaceous and Cenozoic History of North American Vegetation
A complex series of fluctuations have been proposed
that parallel many of the intuitive or anticipated biotic ef-
fects of a bolide impact (Wolfe, 1990). Application of the
CLAMP program to leaf assemblages from western North
America on each side of the boundary has given results in-
terpreted as evidence for profound changes in climate, in-
cluding first a cooling (impact winter) of 1-2 months then
an increase in MAT of 10°C (greenhouse effect; Wolfe,
1990), but there is no evidence for freezing in the form of
frost heaving in paleosols in eastern Montana (Retallack,
1994). A 400% increase in rainfall was also proposed. This
reconstruction is consistent with an earlier suggestion that
because there was no known crater, the bolide may have
fallen in the deep ocean, thus increasing moisture and
seeding the atmosphere with microscopic debris. A crater
of the proper age has now been found on the Yucatan
Peninsula, Mexico, and this complicates the proposed ex-
planation as primary for such a dramatic increase in rain-
fall and coordinated cooling-warming phases. However,
the possible occurrence of hypercanes (runaway hurri-
canes) caused by various events, including an asteroid im-
pact, are theoretically capable of putting large amounts of
water and aerosols into the atmosphere (Emanuel et al.,
1995), and more tropical vegetation did develop in the
Early Paleocene. A distant impact (Chicxulub crater) and
immediately thereafter a closer one (Manson crater; Koe-
berland and Anderson, 1996) has been read from the paleo-
botanical record, and the time of the event has been placed
in early June (Wolfe, 1991; but see Hickey and McWeeney,
1992; Nichols, 1992). The Manson crater is now dated at
-73.8 Ma. As noted by Crowley and North, the 18O record
is ambiguous in its support of the greenhouse model, and
there is even less confirming evidence that there were sub-
sequent long-term trends in climate directly attributable to
a bolide impact.
Although there may have been a very severe environ-
mental disruption at the K-T, and perhaps a CO2 after-
math lasting tens of thousands of years, there does not
seem to be any strong evidence that the impact affected
the long-term evolution of the climate of the earth. The
"mean background climate state" of the latest Maas-
trichtian (-67 Ma, i.e., just before the impact) does not
seem greatly different from the "postaftermath" Early
Paleocene (-64 Ma). (1991, p. 178)
The greatest disturbance in vegetation is evident in flo-
ras from the Western Interior of North America where com-
plete K-T boundary sequences are preserved. The bound-
ary in the Raton Basin is marked by a 2-cm thick clay layer
with anomalously high amounts of iridium and shocked
quartz. This was the first site where a terrestrial iridium
anomaly was reported in conjunction with a palynologi-
cally defined K-T boundary (Orth et al., 1981). Vegetation
in the Raton Basin included plants with small leaf size
(index -34), thick cuticles, and few drip tips and lianas.
The MAT is estimated at ~21-22°C. Above the clay layer,
fern spores appear in abundance (the "fern spike"; Fleming
and Nichols, 1990; Nichols et al., 1986; Tschudy et al.,
1984), followed by an angiosperm recolonization phase
(Wolfe and Upchurch, 1987b) where the leaf-size index of
the angiosperms doubles (70) and many of the leaves have
drip tips and thinner cuticles. This reflects greater rainfall,
estimated at 1500 mm or more, and a paratropical rain for-
est (Fig. 5.12) that persisted through the Early Paleocene.
Mass kills are also interpreted from a section at Teapot
Dome (J. A. Wolfe, personal communication, 1996), and
K-T boundary extinctions are prominent between the
Upper Hell Creek Formation (late Maastrichtian) and the
Ludlow Member of the Fort Union Formation (Paleocene;
Johnson, 1992). A pattern consistent with some effect of a
bolide impact may also be evident south of pi 48° N in the
southeastern United States between the Maastrichtian Mc-
Nairy Sand flora and the lower Paleocene Naborton assem-
blage of Louisiana with the development of a tropical rain
forest resulting from increased precipitation (Fig. 5.12;
Wolfe, 1978). In the Denver Basin a paratropical rain forest
appears with more deciduous dicotyledons, possibly re-
flecting an initial brief lowering of temperature. In the cen-
tral Alberta Coalspur Formation there is a decrease in
diversity (higher extinction levels) between the Maastrich-
tian (abundant aquiloid pollen) and the Early Paleocene
(abundant gymnosperm pollen and no fern spike; Jer-
zykiewicz and Sweet, 1986).
Changes interpreted as resulting from the impact are
also present in the mesothermal vegetation above pi 50° N.
In the Maastrichtian, predominantly evergreen vegetation
occurred up to pi 58° N, and a polar deciduous forest was
further north. By the Early Paleocene the vegetation be-
tween 50° N and 58° N was almost entirely a broad-leaved
deciduous forest (Fig. 5.12; trochodendroids, hamameli-
daleans, Tiliaceae). Evergreen angiosperms were essen-
tially eliminated from North American mesothermal cli-
mates, along with araucarian holdovers. As noted, Wolfe
and Upchurch (1986,1987a; see also Wolfe, 1987) attribute
much of this change to a drop in temperature to near freez-
ing as a result of the bolide impact. Others emphasize al-
ready declining temperatures (Hickey, 1981, 1984) and in-
creasing seasonality in the Late Cretaceous (Batten, 1984;
Frederiksen, 1989a; Srivastava, 1994b; Sweet et al., 1990).
In other regions to the north, in Japan, and in the south-
ern hemisphere K-T sections are not complete and the ef-
fect of the bolide impact appears minor (Askin, 1988; John-
son, 1993; Keller, 1993; Keller et al., 1993). Asian floras
reflect slightly cooling temperatures at the boundary, while
floras in the Hell Creek Formation of Montana and Wy-
oming suggest a warming trend (Johnson and Hickey, 1990).
Members of the northern floras were already adapted to
cool and seasonally dark conditions, and in the southern
hemisphere the effects of the impact apparently did not
subject the floras to freezing temperatures. In these regions
the K-T vegetation shows gradual changes more attribut-
able to climate control. At Bug Creek in eastern Montana

there is no striking difference in pedotypes across the K-T
boundary. Annual rainfall for the Upper Cretaceous Hell
Creek Formation is estimated at 900—1200 mm, while for
the Lower Paleocene Tullock Formation it is also estimated
at 900-1200 mm (Retallack, 1994).
Another aspect of the complex explanation for biotic
changes at the boundary is that the deciduous habit of
northern plants probably rendered them better capable of
surviving and recovering from the impact winter than
broad-leaved evergreen plants whose history is recorded in
relatively great detail in the Raton Basin sequence. It has
been suggested that Cretaceous dinosaurs from the North
Slope of Alaska (pi 70°-85° N) were also adapted to sea-
sonally cold temperatures, as well as to periods of low light
levels, and that the bolide impact was not a major factor in
their extinction there (Brouwers et al., 1987; see also Re-
sponses, 1988).
The methodology used to detect extinctions can influ-
ence interpretations toward either a gradual or a more sud-
den change in diversity. The pollen record, based on iden-
tifications mostly at a level equivalent to biological genera,
reveals extinctions of 20-30% within the region extending
from northern New Mexico to central Alberta. The
megafossil record, where identifications are mostly to a
level equivalent to biological species, show extinctions as
high as 79% in megathermal vegetation (Johnson et al.,
1989), declining to 25% in the polar broad-leaved decidu-
ous forest from central Alberta (Upchurch, 1989; Wolfe and
Upchurch, 1986). Another factor likely accounting for the
differences in the two records is that pollen of the Lau-
raceae does not preserve as fossils, while leaves are promi-
nently represented in megafossil floras of the southern
Rocky Mountains region and are one of the components
showing a major decline in diversity at the boundary. In
contrast, some rare and declining species may be better
and longer represented as pollen, implying a more gradual
rate of extinction. The value of a multifaceted approach is
particularly worthwhile in elucidating any such complex
problem, as noted by Johnson and Hickey with reference to
zonation of K-T boundary sites in the northern Rocky
Mountains and Great Plains: "The fact that the megafloral
zones are not reflected by palynostratigraphy argues for
using an integrated approach to biostratigraphy that com-
bines the high stratigraphic resolution of palynomorphs
with the high taxonomic resolution of megafossils" (1990,
p. 433; see also Lidgard and Crane, 1990). As a general as-
sessment, the overall evidence seems consistent with at
least local to regional biotic changes more rapid than can
be attributed solely to climatic and orogenic trends. This
argues for a bolide impact as one component of vegeta-
tional change at the K-T boundary in the midlatitudes of
North America (Nichols and Fleming, 1990).
A somewhat similar view, but attributing even less long-
term and widespread effect to the bolide, is expressed by
Beeson, who worked on the Cretaceous Corsicana and Ter-
tiary Kincaid K-T boundary formations in Falls County,
Late Cretaceous through Early Eocene History 161
Texas. "The concept of a single catastrophic event as the
operative mechanism for these types of changes is proba-
bly simplistic. It is more reasonable to expect that events
such as bolide impacts only represent a relatively small
contribution to the more significant intrinsic terrestrial
forces acting on biotic change" (1992, p. 18). Hoffman be-
lieves that "The contribution of the environmental con-
sequences of impact and volcanism to the Cretaceous-
Tertiary mass extinction can hardly be disentangled" and
that
It thus appears that the Cretaceous-Tertiary mass ex-
tinction cannot be plausibly interpreted as caused
solely by a bolide impact and its environmental conse-
quences, and it is more feasible to invoke a coincidence
of at least two different factors which were causally un-
related to each other (since the impact occurred at the
end of a period of intense volcanism and extinctions).
(1989, p. 9)
Keller et al. (1993) believe the destructive effects were
negligible in the high latitudes. My own assessment of the
vegetation change at the K-T boundary is that it was, in
part, a consequence of climatic trends already in motion,
which the impact accentuated especially in the midnorth-
ern latitudes of the western hemisphere through alteration
of regional environments.3
An indirect effect of discussions of biotic change at the
K-T boundary has been a reassessment of the precision of
leaf physiognomy in reconstructing past climates. It is be-
coming recognized that older univariate statistics (viz., a
given percent of leaf sizes or entire margins equals a rela-
tively specific rainfall and temperature value) is too sim-
plistic (Wolfe, 1990) and that a number of other factors also
affect leaf form (higher levels of CO2, defense against her-
bivory, etc.). If the 10°C increase in temperature proposed
for just after the K-T boundary is accepted, this must nul-
lify, to some extent, the use of modern southern hemi-
sphere floras as a standard for interpreting Late Cretaceous
and Paleogene leaf margin percentages because these cal-
culations suggest values in the cool megathermal range (for
the Denver and Raton Basins). Also, if the increase is cor-
rect, this alters the previously assumed relationship be-
tween leaf margin percentages and temperature because
these percentages provide a MAT reading of ~10°C at the
K-T boundary as opposed to 23°C derived from CLAMP
ORIGIN AND SPREAD OF DECIDUOUSNESS
Deciduous angiosperms are known from at least the Mid-
dle Cretaceous, while events at the K-T boundary have
been cited as a factor in the spread of deciduousness
among the angiosperms. Three factors have already been
discussed that contributed to its origin and development:
gradual cooling from mid-Cretaceous highs that reached a
threshhold at the K-T boundary, seasonal instability of
streamside habitats, and abrupt lowering of temperatures
162 Late Cretaceous and Cenozoic History of North American Vegetation
resulting from the bolide impact. A fourth factor involves
the migratory history of the angiosperms after their origin
in the early Cretaceous. The group spread from their pre-
sumed region of origin in the tropical uplands of West
Gondwana and radiated into areas that were just fragment-
ing into the various crustal plates. To the north the an-
giosperms encountered the seasonally dry climates of the
late Mesozoic. According to this view, deciduousness orig-
inated on the northern margin of broad-leaved evergreen
vegetation (Axelrod et al., 1991). This preadapted the an-
giosperms to the prolonged winter-dark conditions (a fifth
factor) and the cooler conditions at the high northern lati-
tudes, which they had reached by the Middle- to Late Cre-
taceous, and to the cooling climates of post Late Eocene
times that faciliated their subsequent southern expansion.
With temperature, light, instability of habitats, and
drought already available as long-term factors in the evolu-
tion of deciduousness, it is difficult to ascribe a central role
to the bolide impact with maximum effects of mostly only
a few months' to a few years' duration. Leaf fall is a com-
plex, genetically controlled physiological and anatomical
process that involves the synthesis and functioning of
growth hormones whose effects are coordinated with sea-
sonal changes, formation of abscission layers, dormancy,
and rejuvenation. That the evolution of such processes
would be significantly affected by so brief, albeit dramatic
and intense, an event is difficult to envision. Axelrod et al.
conclude that "The origin and abundance of the deciduous
habit in the Northern Hemisphere, is linked with the re-
striction of seaways, widespread volcanism, and develop-
ment of more continental, cooler climates from the Middle
Paleocene onward, rather than the postulated impact of a
bolide at the close of the Cretaceous . . ." (1991, p. 413).
Regardless of the probable multifaceted, integrated
causes for alteration in climate, the Cretaceous-Paleocene
interval encompasses some important changes in the North
American vegetation. The epicontinental sea was regress-
ing during the Late Maastrichtian (Kauffman, 1977), initi-
ating removal of the barrier between the eastern and west-
ern floristic provinces. Triprojectacites begin to appear
more in eastern North America and Europe, and Norma-
polles are found in increasing numbers in western North
America and Asia. The Late Cretaceous was a time of warm,
subhumid (dry), nearly aseasonal climate over the lower
third of the continent with an estimated annual rainfall of
-1100 mm (drier in the southeast).
In the high northern latitudes, events at the K-T
boundary contributed to the diversification of deciduous
taxa during the Paleocene (e.g., Betulaceae, Fagaceae,
Hamamelidaceae, Juglandaceae; Ulmoideae; Manchester,
1987; Nichols and Ott, 1978). As with the spread of decid-
uousness, it is difficult to determine the relative impor-
tance of the bolide impact and lower threshold tempera-
tures in explaining changes in vegetation evident in the
fossil record. Major trends in biotic history are the result
of endogenic (tectonic-orogenic-volcanic-evolutionary-
migratory) and exogenic (solar-induced global climate)
trends periodically influenced by catastrophic events. As
with any complex system, a multiplicity of factors are usu-
ally involved, and the only explanations to be rejected are
those that force a "mine or thine" alternative.
Paleocene Vegetation: 65-56.5 Ma
Many of the difficulties encountered in reconstructing Late
Cretaceous vegetation and paleoenvironments also apply
to the Paleocene (Figs. 5.1, 5.2, 5.5, 5.12, Table 5.1). Al-
though there are many floras of this age in the Rocky
Mountains region and several along the Atlantic and Gulf
coastal plain, overall the floras are few in number in rela-
tion to the geographic extent (the North American conti-
nent north of Mexico) and the time interval involved (-10
Ma). This precludes making detailed reconstructions for
any one time interval or providing high-resolution histo-
ries from closely spaced floras within individual biotic -
physiographic provinces. Many floras are in the process of
being studied or revised, their age limits use of the modern
analog method, some include only microfossils (Elsik,
1968a,b, 1970), and among these the primary goal is often
stratigraphy and correlation (Frederiksen, 1991b; Pocknall,
1987). For these reasons the history is general and interpre-
tations are continually being refined.
Beyond the complex changes at the Maastrichtian -
Danian boundary (Fig. 5.1), two climatic trends mark the
Paleocene. After an initial lowering in the earliest Pa-
leocene, MAT rose steeply (Fig. 3.1) and there was an in-
crease in precipitation, especially south of pi 48° N. These
changes are evident in the widespread formation of peat,
which was subsequently altered to the Tertiary lignites and
coals that now characterize many parts of the midconti-
nent region. Recall from Chapter 2 (Orogeny section) that
the warm, moist winds that presently bring precipitation
into the southeastern United States developed in the
model simulations with the Paleocene-Eocene uplift of the
Rocky Mountains. The increasing warmth and moisture is
clearly reflected in the floras. Overall diversity increases,
there is rapid diversification of juglandaceous pollen, new
morphotypes appear [Carya, Symplocos, thick-walled
Alangium, Restionaceae, Proxapertites (an extinct relative
of the Old World coastal tropical palm Nipa}], and
Normapolles decline (Frederiksen, 1989b). Leaf size in Pa-
leocene floras is almost always larger than in Late Creta-
ceous assemblages, and half or more of the angiosperm
components in the southern United States floras have drip
tips (Upchurch, 1989). Forests covered the landscape, and
there is little evidence of extensive open country (Axelrod,
1992b). Microthermal vegetation was restricted to the
higher northern latitudes, and in the far north deciduous-
ness included response to low winter light.
Increased precipitation had its greatest effect on the al-
ready warm but dry biota in the southern United States,
while rising temperatures had a significant effect on the al-

ready moist but cool polar communities. In the southern
region an important consequence of greater moisture was
the initial appearance there of a community recognizable
in physiognomy and climatic parameters as a closed-
canopy, multistratal, tropical rain forest (Fig. 5.12). At
Naborton and Mansfield, Louisiana, the estimated MAT for
the Early Paleocene is ~27°C.
Several consequences may be expected from the ap-
pearance and spread of a tropical rain forest (Wolfe and
Upchurch, 1987a). Favorable niches would exist for the de-
velopment of tall trees in response to high precipitation,
luxuriant growth, and dense spacing; buttressing; liana
habit; epiphytic habit; understory habit; and seeds capable
of shade germination. Fossil representatives of families that
include tall canopy-forming trees in extant megathermal
vegetation (Anacardiaceae, Lauraceae, Rutaceae; pollen of
the Bombacaceae-Sterculiaceae-Tiliaceae complex) are
rare or absent in Late Cretaceous vegetation, become in-
creasingly prominent in the Paleocene, and are abundant
and diverse by the Early Eocene. Families that include nu-
merous lianas (Icacinaceae, Menispermaceae, Vitaceae)
show a similar pattern of diversification. Diaspore size in-
creases from the Late Cretaceous to the Paleocene, proba-
bly in response to the larger food reserve needed to com-
pensate for limited seeding photosynthesis in shaded
habitats (Tiffney, 1984). Other adaptations, such as but-
tressing, epiphytism, and understory habit, are difficult to
detect from the fossil record. The appearance of this novel
tropical rain forest formation was a major step in the mod-
ernization of the Earth's vegetation, and its history for
North America north of Mexico begins in the Paleocene.
Associated with the tropical rain forest were present-
day temperate plants that may have lived in the Ap-
palachian highlands. The U.S. Geological Survey's upper
Paleocene Oak Grove core in northern Virginia (Frederik-
sen, 1991a) contains pollen similar to Alnus and Betula, al-
ready present in older Cretaceous deposits, small forms of
Carya, the Alfaroa-Engelhardia-Oreomunnea complex
(presently growing in upland temperate habitats in tropical
latitudes), Platycarya (an eastern Asian temperate tree of
the Juglandaceae; essentially terminal Paleocene), Ilex
(widespread in warm-temperate to subtropical regions),
and the Ulmaceae. A similar type of vegetation is reflected
by the pollen flora of the Paleocene Rockdale lignite from
Milam County, Texas (Elsik, 1968a,b, 1970) that includes
spores referable to the bryophytes (Sphagnum), lycopods,
a diverse fern component, palms, Taxodiaceae, winged
gymnosperm pollen, and numerous angiosperm pollen
types whose biological affinities to family and genus are
unknown. The overall plant diversity trend in the south-
eastern United States was from a low level at the beginning
of the Paleocene, possibly due partly to the aftermath of the
asteroid impact, to higher diversity in the Middle Paleo-
cene. The increase was likely a result of new and less com-
petitive habitats and increased rainfall (Frederiksen, 1994).
Fossil mammalian faunas of the Middle to Late Paleocene
Late Cretaceous through Early Eocene History 163
show more rapid radiation and include Pantodonta, Tae-
niodonta, and Dinocerata, representing browsers probably
of open woodland. The faunal evidence frequently sug-
gests more open vegetation than does the paleobotanical
record (see Chapter 3).
To the far northeast, plant megafossils from the early Pa-
leocene Atanikerdluk flora of Greenland continue to reflect
a polar broad-leaved deciduous forest of trochodendroids,
platanoids, and other Hamamelidae, mixed with decidu-
ous coniferous elements such as Metasequoia (Koch,
1963).4 Some megathermal elements occupied coastal re-
gions and expanded in the later Paleocene and Early
Eocene. The Late Paleocene-Early Eocene Thyra 0 flora
(0= island) from northeastern Greenland (Prinsesse Dag-
mar 0; pi 77°-79° N) consists of 30-31 megafossil species,
including 22-23 angiosperm leaf types. Identifications in-
clude Equisetum, Dennstaedtia, Cephalotaxus, Cupres-
saceae, Elatocladus, Fokienia, Ginkgo, Metasequoia, Betu-
laceae, Cercidiphyllum, cf. Corylus, Platanus, and several
unidentified leaves. Large entire-margined specimens were
assigned to Musophyllum (now Musopsis groenlandicum;
Boyd, 1990, 1992) and are interpreted as an evergreen
monocotyledon of the Heliconiaceae-Musaceae-Strelitzi-
aceae complex. They are suggested as indicating a nearly
frost-free, warm-temperate to subtropical climate along the
coast with a MAT of 15-20°C (Boyd, 1990). This is warmer
than other estimates of 10-15°C for the region north of
65°-70° N based on leaf physiognomy (Wolfe, 1985, 1987)
and 12-15°C (Basinger et al, 1994) and 7.9-9.3 ± 2°C for
Paleocene and Eocene MAT based on multiple regression
analyses (Spicer and Parrish, 1990). Boyd (1990) believes
that leaf physiognomy yields an anomalously low tempera-
ture estimate, at least for the Late Paleocene of northern
Greenland, and Axelrod et al. (1991) suggest seasonal dark-
ness as a factor in the high percentages of lobed leaves.
In the west, fossil wood of Paraphyllanthoxylon abbottii
from the Paleocene Black Peaks Formation of Big Bend Na-
tional Park, Texas (Torrejonian-Tiffanian NALMA or pro-
vincial ages), lacks distinct growth rings (Wheeler, 1991).
In the midcontinent region, the Fort Union Formation (or
Group) of Wyoming, Montana, and North Dakota is an ex-
tensive series of strata ranging in age from Early through
Middle Paleocene. Fossils described by Brown (1962) from
the Fort Union Formation are mostly older than those from
the Sentinel Butte Formation (Crane et al., 1990) and the
Golden Valley Formation (Bear Den and Camels Butte
members; Hickey, 1977). The four floras constitute a series
from the Early Paleocene to the Early Eocene. Brown
(1962) recognized 170 species of plant megafossils, includ-
ing diverse ferns, Ginkgo (Fig. 5.13A), Taxodiaceae (Glyp-
tostrobus, Metasequoia; Fig. 5.13B), araucarias, palms, and
numerous notophyllous angiosperm leaves with serrate
margins and larger entire-margined types. Although many
generic identifications are wrong, there are specimens that
resemble Sabal (Fig. 5.13C), Juglandaceae (Carya, Juglans,
Pterocarya; Cruciptera, Manchester, 1991; Polyptera, Man-
 164 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 5.13. (a) Ginkgo adiantodes from the Paleocene Fort Union Formation, eastern northern Rocky Mountains-
Great Plains region (Brown, 1962, pi. 10, fig. 5). (b) Metasequoia occidental from the Paleocene Fort Union Formation
eastern northern Rocky Mountains-Great Plains region (Brown, 1962, pi. 12, fig. 5). (c) Sabal imperialis from the Pale-
ocene Fort Union Formation, eastern northern Rocky Mountains-Great Plains region (Brown, 1962, pi 14 fig 2) (d)
Platanus raynoldsi from the Paleocene Fort Union Formation, eastern northern Rocky Mountains-Great Plains region
(Brown, 1962, pi. 31, fig. 6). Plant microfossils from the Paleocene-Lower Eocene Fort Union/Wasatch formations Pow-
der River Basin, Wyoming-Montana, (e) Caryapollenites veripites. (f) Momipites wyomingensis. (g) Ulmipollenites
krempii. (hj Almpollemtes verus. (i) IntratriporopoUenites sp. Reprinted from Pocknall (1987) with the permission of
the American Association of Stratigraphic Palynologists Foundation.

Chester and Dilcher, 1967—the oldest unequivocal Juglan-
daceae), Betulaceae (Betula, Corylus; Palaeocarpinus,
Manchester, 1996), Cornaceae (Corpus), Fagaceae (Cas-
tanea, Quercus), platanoids (Fig. 5.13D), Ulmaceae (Celtis,
Ulmus, Zelkova), Moraceae (?; Ficus], Lauraceae (Cin-
namomum, Lindera, Persea, Sassafras), Cercidiphyllum,
Magnolia, Vitaceae, Viburnum^), and others. Ceratophyl-
lum (C. furcatispinum) occurs in the Fort Union Group in
Montana (Herendeen et al., 1990). Pollen assemblages
(Pocknall, 1987) include microfossils resembling Juglan-
daceae (Carya = Caryapollenites, Fig. 5.13E; Platycarya, see
also Wing and Rickey, 1984; the Alfaroa-Engelhardia-Ore-
omunnea complex = Momipites, Fig. 5.13F), Ulmaceae
(Ulmus-Zelkova = Ulmipollenites, Fig. 5.13G), Alnus (=A1-
 Late Cretaceous through Early Eocene History 165

Figure 5.14. Fruits of Trochodendron-Nordenskioldia. (A) Modern fruit of Trochodendron aralioides (original magnifi-
cation x7). (B) Fossil fruit of Nordenskioldia borealis (original magnification x5) from the Paleocene Tongue River Mem-
ber, Fort Union Formation, near Melville, Montana (Crane, 1989, fig. 5c,e). (C) Fossil fruit of Nordenskioldia borealis from
the Paleocene Sentinel Butte Member, Ft. Union Formation, near Almont, North Dakota (Crane, 1989, fig. 7d; Crane et al.,
1990, fig. 6c). Reprinted with permission of Springer-Verlag (Fig. 5.14A-C) and the University of Chicago Press (Fig.
5.14C).

nipollenites, Fig. 5.13H), and the Bombacaceae-Sterculi-
aceae-Tiliaceae complex (Intratriporopollenites, Fig. 5.131).
In general, localities south of about central Wyoming in-
clude Cinnamomum, Menispermum, and palms; those
north of the Colorado-Wyoming border include Ginkgo,
Acer, Betula, Corylus, Viburnum, and other temperate gen-
era. Physiographic relief was moderate, but recall from
Chapter 2 that nearly horizontal subduction of the Farallon
Plate extended as far east as the Black Hills and produced
uplands in the region. Also, in a temporal sense the Fort
Union vegetation was growing during the transition be-
tween cooler Late Cretaceous-earliest Paleocene tempera-
tures and the onset of warmer conditions. In a geographic
sense the region was near the ecotone between the north-
ern microthermal polar broad-leaved deciduous forest and
the megathermal dry tropical forest-evergreen tropical
rain forest expanding from the south (Fig. 5.12). Wolfe
(1985) interprets the flora of the Tongue River member of
the Fort Union Group (Powder River Basin) as a paratropi-
cal rain forest, while Wing (1981) believes the slightly
younger early Willwood flora of the Big Horn Basin repre-
sents a notophyllous broad-leaved evergreen forest.
The Sentinel Butte flora near Almont in central North
Dakota is Late Paleocene in age. It presently comprises
30-35 species, including Ginkgo adiantoides (Ginkgo-
aceae), cf. Parataxodium (Taxodiaceae), cf. Canticocculus
(Menispermaceae), Cornus (Cornaceae), Cyclocarya brownii
(Juglandaceae; Manchester, 1987; Manchester and Dilcher,
1982), Meliosma rostellata (Meliosmaceae-Sabiaceae), Nor-
denskioldia borealis (an extinct genus of the Trochoden-
draceae; Crane et al., 1990; Fig. 5.14A-C), Nyssidium-Joff-
rea complex (Cercidiphyllaceae; Crane, 1989; Fig. 5.ISA,
B), Paleomyrtinaea princetonensis (Myrtaceae; Pigg et al.,
1993), Paleocarpinus (Betulaceae), Palaeophytocrene (Ica-
cinaceae), Platanaceae, Psidium (Myrtaceae), Spiremato-
spermum (Zingiberaceae; Manchester and Kress, 1993),
and others whose familial affinities are uncertain (e.g.,
Porosia verrucosa; araceous?).
The Golden Valley Formation in the Williston Basin of
western North Dakota provides a sampling of the eastern-
most Paleocene and Eocene floras of the Rocky Mountain
region (Hickey, 1977). It consists of the Bear Den Member
with 41 plant species of latest Paleocene age (56 Ma) and
the Camels Butte Member with 37 plant species of Early
Eocene age (53-55 Ma). The most abundant fossils identi-
fied from the former unit are Metasequoia (Taxodiaceae),
Corylus (probably Paleocarpinus; Betulaceae), "Meliosma"
(probably a platanoid), "Cocculus" (Menispermaceae; now
Nordenskioldia, Manchester et al., 1991), Cercidiphyllum,
and Chaetoptelea (Ulmaceae); they represent a lowland
forest. Important associates include Platanus and extinct
members of the Menispermaceae [Menispermites], Rosidae
(Averrhoites), Hamamelidaceae ("Viburnum" cupanoides),
and a laurel close to the modern genus Persea (L. J. Hickey,
personal communication, 1996). Owing to the develop-
ment of an extensive shallow lake in the earliest Eocene,
swampy and aquatic vegetation, including Equisetum,
Glyptostrobus (Taxodiaceae), Sparganium (Sparganiaceae),
and Porosia (a floating aquatic) become more important to-
ward the top of the Bear Den Member. The Bear Den vege-
tation shows closest affinity with the immediately older
Paleocene floras from the region. It is a notophyllous
broad-leaved evergreen forest mixed with some deciduous
species.
The woody flora of the slightly younger, Early Eocene,
Camels Butte Member is dominated by Platycarya ameri-
cana, "Meliosma" (Platanaceae), and Ternstroemites (Thea-
ceae), associated with Averrhoites, Dombeya (Sterculiaceae),
166 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 5.15. Fossil leaf of
Cercidiphyllaceae (Trocho
dendroidesarctica-Nyssidium-
Joffrea complex) from the
Paleocene Tongue River
Member, Fort Union Forma-
tion, near Melville, Montana
(Crane, 1989, fig. 8b,c).
(A) original magnification xl.
(B) original magnification x3.
Reprinted with permission
of Springer-Verlag.
Metasequoia, and other extinct Theaceae (Paraternstro-
emia). The floating fern Salvinia preauriculata is the most
conspicuous element of the swamp and aquatic flora of the
Camels Butte Member; all of the paludal and lacustrine Pa-
leocene forms except Osmunda persist, and these are
joined by Isoetites (Isoetaceae), Lygodium (Schizeaceae),
Nelumbo, and Zingiberopsis (Zingiberaceae). The Camels
Butte flora shares most of its species with other Eocene flo-
ras and, in general, shows little similarity to the immedi-
ately older Paleocene floras. [For a complete listing of
plants identified from the Golden Valley Formation see
Hickey (1977) with the revisions and additions noted
above.1
The Willwood flora from the Bighorn Basin of north-
western Wyoming is of latest Paleocene through Early
Eocene age (52.8 ± 0.3 Ma; earliest lost Cabinian/earliest
(Wasatchian NALMA or provincial age) and shows a pro-
gressive increase in entire-margined leaves (41-52%). This
reflects a change from a deciduous forest in the Late Paleo-
cene toward a mixed evergreen and deciduous broad-
leaved forest with seasonal rainfall, but no widespread
aridity, and nearly frost-free conditions (Wing, 1980, 1981,
1984, 1987; Wing and Bown, 1985; Wing et al., 1991; Table
5.4; Fig. 5.12). The MAT is estimated at a maximum near
18°C and a cold-month average above 0°C (probably above
5°C) for the late Early Eocene. Evidence from leaf physiog-
nomy suggests a cold-month mean temperature of 1-8°C.
Computer models for the continental interior at -50 Ma
hindcast harsh winters, but the widespread and consistent
paleobotanical and faunal evidence indicate a need for re-
vision in these sensitivity tests. The Lower Eocene part of
the Willwood Formation contains an extensive fossil mam-
malian fauna presently under study (crocodile relatives,
land turtles, tree-dwelling mammals; Bown et al., 1994)
and, together with the contemporaneous flora, an extensive
data base is becoming available for reconstructing the biota
and paleoenvironments.
The Clark's Fork Basin floras range through the Puercan
(Early Paleocene), Tiffanian (latest Paleocene), and Clark-
forkian (latest Eocene) NALMA or provincial ages. The re-
ported composition is similar to that of other Paleocene
floras in the region (Hickey, 1980; S. L. Wing, personal
communication, 1996): Ginkgo, Metasequoia, Glyptostro-
bus, Cercidiphyllum, Corylus, Juglandaceae, Meliosma
(=Platanaceae), Platanus, and others. The Tiffanian MAT at
Clark's Fork is estimated at 10°C and the Clarkfordian MAT
at Clark's Fork at 13.5°C (Hickey, 1980).
The stratigraphic relationship between floras in the iso-
lated basins of the Great Plains is complex (Fig. 5.16). A
summary of paleoclimatic estimates for the Sentinel Butte,
Bear Den, Camels Butte, Clarks Fork, and other assem-
blages in the upper Great Plains has been provided by Crane
et al. (1990) and Wing et al. (1991). The Sentinel Butte flora
contains few if any broad-leaved evergreens, only one
conifer (cf. Parataxodium), and is dominated by deciduous
plants indicating a temperature-seasonal climate. The most
similar modern analog is the mixed mesophytic forest of
Asia and vegetation along the south part of the Atlantic
coastal plain of eastern North America. Entire-margined
leaves constitute 45% of the flora, which is higher than in a
mixed mesophytic forest and lower than in a tropical or
paratropical forest. The closest matching figures from data
plotted by Dolph and Dilcher (1979) are from the piedmont
of North and South Carolina, where the MAT is 15°C, the
coldest month average is 4.5°C, and the warmest month av-
erage is 24°C. For Bear Den time the MAT is estimated from
13°C at 56 Ma to 15-16° C at -55 Ma and at 17.8°C for the
Camels Butte flora (Hickey, 1977). The MAT for the Powder
River Basin in northern Wyoming is estimated at 16-18°C
between 54 and 50 Ma (Wing et al., 1991) and 18-21°C for
the Raton Basin further south at the New Mexico-Col-
orado boundary (Wolfe, 1990; Wolfe and Upchurch,
1987b).
If these regional MAT estimates are placed in sequence
(cf. Fig. 5.16), the values are 10°C (Tiffanian), 13.5°C (Clark-
fordian), 15°C (Sentinel Butte), 13°C (Bear Den, 56 Ma),
15-16° C (Bear Den, 55 Ma), and 17.8°C (Camels Butte,
55-53 Ma). In general, they show recovery from Early Pa-
leocene lows and the trend toward the Cenozoic thermal
maximum reached in the Early Eocene. However, the trend
was not an uninterrupted progression from cooler to
warmer conditions. Wing (1996) found that high-resolution
 Late Cretaceous through Early Eocene History 167

Table 5.4. Plant megafossils from Early Eocene part of Willwood Formation, Bighorn Basin, northwestern Wyoming.
Pond Swamp Levee/Crevasse Splay
Zingiberopsis Zingiberopsis Zingiberopsis
Menispermites Menispermites Menispermites
Amesoneuron (palm leaf fragment) Amesoneuron Amesoneuron
Platanus Platanus
Salvinia Salvinia
Betulaceous leaves Betulaceous leaves
Cercidiphyllum Cercidiphyllum
Populus Populus
?Typhaceae ?Typhaceae
Equisetum Equisetum
Allantoidopsis Allantoidopsis
Alnus Alnus
Averrhoites Averrhoites
Glyptostrobus Glyptostrobus
Leguminosites Nemaleison Meliosma
Salix Lygodium lAmpelopsis
Spirodela Cnemidaria aff. Carya
" Sparganium" (Cyperaceae) Phoebe Woodwardia
?Rutaceae Platycarya ?Tiliaceae
?Berberidaceae Dombeya Corn us
IDalbergia Thelypteris fEugenia
Potamogeton "Sequoia" Fagopsis
Leguminosae ?Magnoliaceae Ginkgo
Menispermaceae " Apocyn ophyllum " Metasequoia
Azolla Ulmus
(Chaetoptelea)
Placement unknown
Penosphyllum
Persites
Polyptera
tSchoepfia
Adapted from Bown et al. (1994), Farley (1990), and Wing (1981, 1984).

sampling, combined with critical taxonomic studies of fos-
sil floras in the northern Rocky Mountains region, reveals
a cool excursion of 1-1.5 m.y. at the Paleocene-Eocene
transition. This important work brings paleobotanical data
into closer agreement with emerging theoretical consider-
ations about the nature of climatic change. It is becoming
increasingly evident that trends in climate are character-
ized by reversals and abrupt threshold changes. These are
frequently obscured in older floras by the uneveness of the
fossil record and broad sampling intervals. Another inter-
esting new result is that fossil faunas in the region reflect
the same cool interval (K. D. Rose, personal communica-
tion, 1996), further demonstrating the value of synergistic
approaches to paleoenvironmental studies and the impor-
tance of context information.
Faunal evidence also favors progressively warmer MAT
(Hutchison, 1982). Koch et al. (1992; see also Rea et al.,
1990) have studied the isotopic carbon ratios in tooth
enamel of Phenacodus and paleosol carbonates from the
Fort Union (65-55 Ma) and Willwood formations (55-52
Ma) in the Bighorn Basin of northwestern Wyoming, which
spans the Paleocene-Eocene boundary. A decrease in 13C
occurs across the boundary and probably reflects increas-
ing temperatures. Paleocene vegetation was mesothermal-
megathermal broad-leaved deciduous forest giving way to
more megathermal evergreen tropical vegetation later in
the Paleocene and Early Eocene (cf. Fig. 3.1), as indicated
by such frost-sensitive indicators as the tree fern Cnemi-
daria, palms, Zingiberaceae, and cycads.
Pabst (1968) described the sphenophytes, ferns, and
gymnosperms from the Paleogene (Eocene?) Chuckanut
Formation of northwestern Washington (latitude -48° N),
which are generally similar to Paleocene floras previously
described (Equisetum, numerous ferns, Glyptostrobus,
Metasequoia, Taxodium).
North of about pi 50° the earliest Paleocene assemblages
have leaf-margin percentages indicative of a microthermal
climate with possible January temperatures near freezing.
A flora from the Paleocene Ravenscrag Formation in south-
western Saskatchewan, Canada grew on a broad alluvial
plain with meandering streams, ponds, and swamps. The
vegetation included a broad-leaved deciduous forest (Pla-
tanites canadensis; Trochodendroides speciosa, Fig. 5.17A;
Amelanchites similis; Cornophyllum newberryi; Dicotylo-
phyllum anomalum; Ginkgo), needle-leaved deciduous
forest (Fokienia ravenscragensis, Fig. 5.17B; Glyptostrobus
168 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 5.16. Stratigraphic relationship between Upper Cretaceous and Lower Tertiary formations of the midcontinent re-
gion. Modified from Mclver and Basinger (1993, fig. 4) and based on references cited therein. Reprinted with the permis-
sion of the Canadian Society of Petroleum Geologists.

nordenskioldii; Metasequoia occidentalis], with a few
needle-leaved evergreens and possibly a rare broad-leaved
evergreen (Myexiophyllum americanum). The arrangement
of the principal species are shown in Fig. 5.18, and the
MAT is estimated at 16°C. The Joffre Bridge flora (latitude
-65° N) near Red Deer in Alberta, Canada, is from the
Paskapoo Formation of Late Paleocene (Late Tiffanian) age
(Bell, 1949). It is presently under study but known to con-
tain Botrychium (Rothwell and Stockey, 1989), Onoclea
sensibilis (Rothwell and Stockey, 1991), the Cercidiphyl-
lum-like Joffrea speirsii (Crane and Stockey, 1985), Limno-
biophyllum (Lemnaceae; Stockey et al., 1997), Paleocarpi-
nus joffrensis (Betulaceae, aff. Corylus; Sun and Stockey,
1992), platanoid remains [various parts described as Pla-
tanus (or Macginitiea; leaves), Macginicarpa (pistillate in-
florescences), and Platananthus (staminate inflorescences;
Pigg and Stockey, 1991)], and abundant Azolla (Hoffman
and Stockey, 1992). Associated remains include Equise-
tum, Osmunda, Woodwardia, Ginkgo, Glyptostrobus,
Metasequoia, ^leer-like fruits, Chaetopetala, Porosia, and
Viburnum-like leaves; deciduous Pinaceae (Pseudolarix)
are present in other related floras.
A Late Paleocene palynoflora from the Eureka Sound
Group, Northwest Territories, includes Sphagnum, Ly-
copodium, Osmunda, Schizaea, Cupressaceae-Taxodiaceae,
Picea, Pinus, Sparganium, cf. Acer, Alnus, Betula, Corylus,
Ericaceae, Juglandaceae (Momipites-Caryopollenites, Pte-
rocarya), Liquidambar, Pachysandra, cf. Quercus, Tilia,
and Ulmus (Mclntyre, 1989). Absent from the region are
palms and broad-leaved evergreens and entire-margined
species referred to Cinnamomum, Ficus, Magnolia, and
Persea that were so abundant to the south. In the Bonnet
Plume Formation of the Yukon Territory, Rouse and Srivas-
tava (1972) note a floristic change across the K-T bound-
ary, indicating cooler Paleocene conditions. Angiosperms
(Alnus, Betula, Corylus, Fraxinus, Myrica, Ulmus, tro-
chodendroids, platanoids and other hamamelids) and
conifers (Cedrus, Glyptostrobus, Metasequoia, Picea, Pinus,
Taxodium) were apparently codominant at the high north-
ern latitudes. These shifted further north and to higher el-
evations as temperatures began to rise in the Paleocene.
The Late Paleocene (53.3±1.5to55.8±1.7 Ma; Triplehorn
et al., 1984) Chickaloon flora of Alaska is a broad-leaved
evergreen and coniferous forest of Taxodiaceae (Glyp-
tostrobus, Metasequoia), trochodendroids (Nordenski-
oldia, Joffrea), and several members of uncertain affinities

("Carya" antiquora, Averrhoites affinis, Macginitiea, "Grew-
iopsis" sp., "Meliosma" longifolia, Fagopsis groenlandica;
Wolfe, 1994). Broad-leaved evergreen vegetation eventu-
ally extended to -60° N in the Early Eocene. As noted pre-
viously, Wolfe (1985, 1987) estimates a MAT of 10-15°C
(CLAMP 12.3°C) poleward from -65° N to 70° N, while
Boyd (1990, 1992) places the range at a seemingly warm
15-20°C in northernmost Greenland to account for Musop-
sis (Musophyllum).
The number of extant angiosperm families doubles from
that of the Maastrichtian based on fossil pollen evidence.
Prominent families in the Paleocene are the Betulaceae
(only Alnus and possibly Betula in the Late Cretaceous),
Fagaceae, Hamamelidaceae, Ulmoideae, and Juglandaceae
(Manchester, 1987), which diversified during this period.
Lianas probably assignable to the Tiliaceae (Grewia type)
also appear. Conifers at the lower latitudes are generally
rare.
Evidence for herbaceous vegetation during the period
from the Maastrichtian through the Paleocene is meager in
both the micro- and megafossil record. Spores resembling
the Blechnaceae-Polypodiaceae-Pteridaceae complex are
represented and megaspores of the lycopods, Gleicheni-
aceae, and Schizaeaceae are present. Crane (1987) suggests
that lycopods, ferns, and angiosperm shrubs played a
prominent role in understory and colonizing vegetation
during the Late Cretaceous and Paleocene. [See Spicer et
al. (1985) for a discussion of Pityrogramma calomelanos as
the primary colonizer on Volcan Chichonal, Chiapas, Mex-
ico).] As a generalization, the Paleocene vegetation was rel-
atively uniform across the Holarctic region and typically
included Ginkgo, Metasequoia, Glyptostrobus, MacGini-
tiea and other members of the Platanus complex, early rep-
resentatives of the Carya and Ampelopsis lineages, and
Late Cretaceous through Early Eocene History 169
Figure 5.17. (A) Leaf of
Trochodendroides speciosa
from the Paleocene Ravenscrag
Formation, southwestern
Saskatchewan, Canada.
Reprinted from Mclver and
Basinger (1993, pi. 21, fig. 1)
with the permission of the
Canadian Society of Petroleum
Geolgists. (B) Leafy branch of
Fokienia ravenscragensis from
the Paleocene Ravenscrag
Formation, southwestern
Saskatchewan, Canada.
Reprinted from Mclver and
Basinger (1993, pi. 13, fig. 2)
with the permission of the
Canadian Society of Petroleum
Geologists.
various members of the Cercidiphyllum complex (Wing,
1987). In a generalized south to north direction, Paleocene
vegetation consisted of broad-leaved tropical rain forest
(Gulf Coast), warm-temperate broad-leaved evergreen for-
est with few deciduous hardwoods (Raton, Denver floras),
mixed deciduous and evergreen broad-leaved forest north
of Colorado-Wyoming, and mixed deciduous-coniferous
forests with a few montane conifers further north (Axelrod
et al., 1991). Later in the Eocene many of the present-day
Asian genera were eliminated from the eastern Rocky
Mountains and central Plains region because of seasonal
dryness. These floras will show greater affinities with the
modern vegetation of Mexico and Central America in a
complex pattern that may have begun in the latest Paleo-
cene and Early Eocene.
Early Eocene Vegetation: 56.5-50 Ma
During the early Eocene temperatures continued to rise,
reached their highest values in all of Cenozoic time at
50-55 Ma, and persisted at these values for 2-5 Ma (Koch
et al., 1992) (Figs. 5.1, 5.2, 5.7, 5.19). The temperature gra-
dient for the continental interior is estimated at ~C.4°C/1°
latitude (Greenwood and Wing, 1995) and was probably
less along the coasts. Rainfall also continued at high levels.
An increase in volcanic activity in the Late Paleocene at
-56 Ma identifies CO2 as a factor in the warming trend
(Vogt, 1979; see also Berner et al., 1983; Owen and Rea,
1985; Sloan and Rea, 1995). Another warm period at 17-15
Ma (Fig. 3.1), also corresponding to a peak in volcanic ac-
tivity, further supports an interaction between volcanism,
CO2, and temperature. There is a marked decrease in the
ocean carbon reservoir (increasing temperatures). The
Norwegian-Greenland Sea was opening at this time and
170 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 5.18. Profile diagram of the Ravenscrag Butte Paleocene forest community. Reprinted from Mclver and Basinger
(1993, fig. 5) with the permission of the Canadian Society of Petroleum Geologists.
the Thulean volcanic episode was in full force. An abrupt
decrease in the size of eolian particles reflects moist condi-
tions and more dense vegetation cover (Rea et al., 1985).
Megathermal tropical rain forest vegetation attained its
greatest latitudinal extent, reaching to near 45°-50° N (fur-
ther north along coasts), and lateritic paleosols are found at
the same latitudes. Paratropical rain forest extended to
60°-65° N, and fossils of a meso- to microthermal noto-
phyllous broad-leaved evergreen forest occur on an exotic
terrane at 70°-75° N along the northeast Pacific (Wolfe,
1972, 1977, 1985), or (Axelrod et al., 1991) a polar broad-
leaved deciduous forest occupied the continental interior.
Several taxa are found in North America for the first time
during this interval. The first appearnace of Platycarya ap-
proximately marks the Paleocene-Eocene transition, and
megafossils of the ferns Cnemidaria magna and Lygodium
kaulfussi indicate an Eocene age for the interior basins
(Bown et al., 1994). Leaves of Quercus are found in the
Early Eocene of the Sierra Nevada (MacGinitie, 1941). The
Paleocene-Eocene transition was also marked by extinc-
tions in both plant and mammalian groups, and overall
there was a decrease in diversity (Wing, 1997; Wing et al.,
1995).
The Eocene history of vegetation in the southeastern
United States is difficult to trace because the floras are cur-
rently being revised on a paleomonographic basis and their
composition is not fully known. The floras were initially
studied by Berry (1916, 1922, 1924, 1930, 1941), but
Dilcher (1971,1973) has found that -60% of the identifica-
tions are incorrect. The fossil reported as Taxodium is a
Podocarpus; the family Proteaceae, represented by four
genera fide Berry, is not present; eight of 10 species of
Sapindus are one form and it is not Sapindus; and one
form described as Aralia is Dendropanax. Also, the dating
of the fossil-bearing strata has had a complex history. Berry
thought that his specimens came from the Wilcox Forma-
tion, dated as latest Paleocene to Early Eocene, and repre-
sented a coastal strand depositional environment. The
most extensive fossil-bearing deposits are from the Bell
City pottery clay pit near Bell City in western Kentucky
and from the Puryear clay pit near Puryear, Henry County,
Tennessee. The fossils at these sites are now assigned to the
Claiborne Formation of Middle Eocene age, dated primar-
ily on the basis of palynostratigraphy, and were deposited
in ox-bow lakes and channel fills of meandering rivers sev-
eral miles inland from the coast. Furthermore, until re-
cently other plant-bearing deposits from western Ten-
nessee (e.g., the Buchanan clay pit) were thought to be in
the Paleocene part of the Wilcox. These deposits are now
dated as Early Eocene Wilcox (N. O. Frederiksen, personal
communication, 1992), which affects by some 5 m.y. re-
ports of the early to earliest fossil record of various taxa
(e.g., grasses; Crepet and Feldman, 1991; see also Crepet
and Taylor, 1985,1986). The relationship between stages of
the Paleocene through the Oligocene and Gulf Coast
provincial stages are shown in the right column of Fig.
5.20.
Floras of the Lower Eocene Wilcox Formation, as
presently defined, have not been extensively studied. Re-
ports from western Tennessee include the grass spikelets
and inflorescence fragments previously mentioned; the
legume Protomimosoidea buchananensis (Fig. 5.21), inter-

EOCENE TIME SCALE
Figure 5.19. Eocene time scale. Reprinted from Berggren et al. (1995) with the permission of the Society for Sedi-
mentary Geology.
preted as combining features of the subfamily Mimo-
soideae and the Dimorphandm group of the subfamily
Caesalpinioideae (Crepet and Taylor, 1985,1986); Barneby-
anthus buchananensis, an extinct genus of the papilionoid
tribe Sophoreae (Crepet and Herendeen, 1992); Pistillipol-
lenites macgregorii (Gentianaceae fide Crepet and Dagh-
lian, 1981, but see Stockey and Manchester, 1988); and var-
ious unidentified mesophyllous, entire-margined leaves
suggesting a megathermal climate and tropical rain forest
vegetation. The extinct Dryophyllum occurs in many Paleo-
Latecreaeouestroughy ecne hitroy 171
gene floras in the southeast and is possibly the complex
from which other Fagaceae were derived (Jones and Dilcher,
1988). It ranges from about the Early Eocene into the Late
Eocene.
Palynofloras from the southeastern United States record a
decline in plant diversity at the Paleocene-Eocene bound-
ary (upper NP9-lower NP10), followed by a sharp recovery
and a significant turnover in vegetation during the Early
Eocene. Numerous Paleocene and earliest Eocene an-
giosperm pollen types disappear and are replaced by new
172 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 5.20. Age and stage equivalency of Paleo-
gene strata in the Gulf Coast region. Note that the
Eocene-Oligocene boundary is now placed at 34
Ma (Chapter 6). Modified from Axelrod (1992b)
and based on Harland et al. (1990). Reprinted
with the permission of Cambridge University
Press.
forms (Frederiksen, 1988). This does not seem attributable
solely to climate, emigration, extinction, or evolutionary
change. Rather, it involved immigration of new elements
over the North Atlantic land bridge. According to Frederik-
sen, "It now appears that the early Eocene event is the most
distinct floral immigration event documented for the Eocene
of North America." (1988, p. 44). Platycarya (Juglandaceae)
and Eucommia (Eucommiaceae) migrated from Europe into
North America (but see Call and Dilcher, 1997). They mostly
became extinct in the United States and Canada in the Late
Eocene-Early Oligocene, persisted in Central Mexico
Figure 5.21. Protomimosoidea buch-
ananensis from the Lower Eocene Wil-
cox Formation, Buchanan, Tennessee.
Reprinted from Crepet and Taylor,
1986, fig. 4) with the permission of the
Botanical Society of America.
(Puebla) into the late Cenozoic (Magallon-Puebla and Ceval-
los-Ferriz, 1994), and are now native only to eastern Asia.
This parallels the strong North American-European affinity
seen in Lower Eocene faunas and documents that the North
Atlantic region posed no serious physical or climatic barrier
to migration. Eocene faunas from Ellesmere Island (Dawson
et al., 1976; West and Dawson, 1978) contain alligator and
arboreal prosimian primates, including an extinct der-
mopteran distantly related to the modern flying foxes
(lemurs) of the Old World tropics (Estes and Hutchison,
1980). The locality is presently at latitude 78° N and the pa-
leolatitude was near 79° N (McKenna, 1980). Thus, very lit-
tle of the change in environment and biota can be attributed
to plate movement. The climates at these highest northern
latitudes supported both megathermal (southern coastal
Thulean route) and more temperate (northern inland
DeGeer route) vegetation, allowing an array of plants and an-
imals to move into the southern United States. Early Eocene
warmth is widely documented in the northern hemisphere,
as indicated by the London Clay flora of England. The MAT
there is estimated at 25°C (77°F) compared to the present
10°C (50°F). The coldest month mean temperature for the
Eocene southern Arctic was 10°C, and limited frost was pos-
sible (Tiffney, 1994).
In the central Rocky Mountains region the Paleocene
and Early Eocene was a time of intense deformation. Most
of the highlands and major basins were present by the
Early Eocene, reaching one-half or more of their present
average elevation. New estimates based on paleobotanical
evidence suggest that, at least locally, elevations may have
 Late Cretaceous through Early Eocene History 173

Fig. 5.22 Major mountain ranges and basins in the central Rocky Mountains region
during the Paleocene and Early Eocene (Leopold and MacGinitie, 1972).

been near modern values (Chapter 6, Paleoelevation Analy-
sis). Numerous lakes formed by the blockage of drainage
systems occupied the basins (Fig. 5. 22), creating favorable
conditions for the preservation of extensive floras and fau-
nas. During the latest Eocene-earliest Oligocene these
large basin lakes disappeared (Franczyk et al., 1992).
In general, the Early Eocene floras of the region repre-
sent mesothermal sclerophyllous broad-leaved evergreen
(tropical to paratropical rain forest) and mixed mesophytic
elements with distinct Asian affinities: Aleurites (Euphor-
biaceae), Dipteronia (Aceraceae), Euptelea (Eupteleaceae),
and Platycarya (Juglandaceae; Leopold and MacGinitie,
1972). The flora of the late Early Eocene Wind River Forma-
tion (-50-51 Ma) has not been studied in detail, but
Leopold and MacGinitie (1972) provide a preliminary list
that includes southeast Asian genera (Actinidia, Alangium,
Cinnamomum, Euptelea, Eustigma, Firmiana, Limacia, Lit-
sea, Machilus, Maesa, Millettia, Neolitsea, Paliurus, Platy-
carya, Pterocarya), a few genera now present in tropical
northern Latin America (Cedrela, Luehea, Mabea, Ore-
opanax, Serjania, Thouinia, Ungnadia), and a number oc-
curring in both regions. The plants suggest humid subtrop-
ical conditions. A few microthermal temperate elements
were present (Acer, Alnus, Populus], but these are not nu-
merous or diverse. In contrast, floras of similar to slightly
younger age west of the Rocky Mountains are often domi-
nated in both diversity and abundance by microthermal
taxa (Copper Basin, Republic, Thunder Mountain; Chapter
6), probably due to the higher elevation (Wing, 1987) and
unimpeded moist winds from the Pacific.
174 Late Cretaceous and Cenozoic History of North American Vegetation
The Wind River Formation also preserves one of the
most diverse Early Eocene vertebrate faunas in North
America. More than 100 species including mammals (65
species), reptiles (22 lizards, three snakes, two turtles);
crocodylians (two), birds (two), amphibians (four), and
fishes (three) have been identified. [Recent parts in the
published series are Dawson et al. (1990), Rose et al.
(1991), and Stucky and Krishtalka (1990).] The fossils are
from the Lost Cabin Member of the Wind River formation
in the latest Wasatchian NALMA or provincial age (50.5
Ma). The fauna represents communities living along flood
plains, small ponds, and well-drained swamps. The lithol-
ogy of the sediments shows evidence of some periodic but
not extensive drying (e.g., filled mud cracks). Species di-
versity and composition suggest a humid, multistoried
tropical to subtropical gallery forest with only slight sea-
sonal differences in rainfall (Stucky et al., 1990). This is
consistent with the more limited paleobotanical data. Fos-
sil plants from slightly younger deposits are known from
Yellowstone National Park (50-51 Ma, early Bridgerian
NALMA), including the famous silicified forests at Speci-
men Ridge and Amethyst Mountain (Dorf, 1960; Knowlton,
1899; Wheeler et al., 1977, 1978); but these have not been
studied in detail. [For a review of the Yellowstone floras see
Wing (1987).] Computer modeling of Early Eocene climates
in the midcontinent region predicted a cold-month mean
temperature 3-10°C below freezing, or about the same as at
present. Increasing the temperatures via a greater CO2 con-
centration for areas such as Wyoming and the Dakotas
overheats the globe generally in the model predictions. In
contrast, a wealth of paleontological evidence from plants
to crocodiles indicates winter temperatures above 5°C
(1-8°C) in Wyoming and Montana (Wing and Greenwood,
1993). Efforts are underway to reconcile the differences
through simplified models that more efficiently carry heat
from the equator and incorporate the effect of large lakes in
the region (Chapter 6, Middle through Late Eocene Vegeta-
tion, Green River flora). This is an example of the impor-
tant role of paleontological verification data in improving
the accuracy of modeling results.
The Middle Eocene Chalk Bluffs flora (lone Formation,
49-52 Ma, Wolfe 1981b; 50-52 Ma, Wing and Greenwood,
1993; -48 Ma, J. A. Wolfe, personal communication, 1996)
occurs along the west slope of the Sierra Nevada in Califor-
nia. It is tropical in aspect, but it has not been revised since
MacGinitie's (1941) original study. Generally, the lowland
vegetation in the Pacific Northwest was megathermal rain
forest.
To the far north, palynofloras are known from the Early
Eocene part of the Eureka Sound Group of Cornwall and
Amund Ringnes islands (Sverdrup Group), Northwest Ter-
ritories. The fossiliferous deposits were initially dated as
Paleocene-Eocene, but they are now regarded as Early
Eocene (Kalgutkar and Mclntyre, 1991; Mclntyre and Rick-
etts, 1989). Identifications include Sphagnum, Cupres-
saceae-Taxodiaceae, Picea, Pinus, Alnus, Betula, Carya,
Corylus, Engelhardia [Momipites], Ericaceae, Ilex, Pachy-
sandra, Pterocarya, Tilia (rare), and Ulmus. Late Paleo-
cene-Early Eocene palynofloras are also known from the
Eureka Sound Group (Iceberg Bay Formation) of Ellesmere
Island (Kalkreuth et al., 1993). These add Glyptostrobus,
Metasequoia, Sequoia, Taxodium, Acer, Cercidiphyllum,
Diervilla, Fraxinus, Liquidambar, Nyssa, Quercus, and
Viburnum. Megafossils from the Eocene Buchanan Lake
Formation, Axel Heiberg Island in the Canadian high Arc-
tic, are from swamp-coal, fluvio-lacustrine shale, and
channel-sand lithofaces (Greenwood and Basinger, 1994).
The vegetation was a mosaic of Metasequoia and Glyp-
tostrobus (taxodiaceous swamp), a mixed coniferous com-
munity (Pseudolarix, LePage and Basinger, 1995a;
fChamaecyparis; Larix, LePage and Basinger 1991a,b,
1995b; Pinus}, Alnus-fem (Osmunda) bogs, and various
broad-leaved deciduous gymnosperms (Ginkgo) and an-
giosperms (cf. Betula, aff. Quercus, cf. Zelkova, Juglan-
daceae, platanoids, Cercidiphyllum complex).
To the northwest, late Early Eocene floras (originally
dated as Middle Eocene) are preserved within the Yakutat
block of Alaska (62° N; e.g., Kushtaka flora; Wolfe, 1977,
1985). The coastal localities include floras with broad-
leaved evergreens, 65% have entire margins, drip tips are
common, and lianas are well represented. Recall that this is
the time of maximum warmth in the Cenozoic and the pe-
riod when megathermal tropical vegetation attained its
northernmost extent. The floras include many plants of trop-
ical Asian affinity identified as Alangium, Alnus, Anamirta,
Barringtonia (a tropical Pacific beach plant), Cananga,
Clerodendrum, Kandelia (a mangrove), Lauraceae (five
species), Luvunga, Limacia, Melanorrhoea, Meliosma,
Menispermaceae (five paleotropical genera), Myristica, the
dipterocarp Parashorea, Phytocrene, Platycarya, Pyrenan-
cantha, Sageretia, Saurauia, Stemonurus, and Tetracentron.
Paleoclimatic conditions are difficult to assess. At the
warmest end of the range the fossils along coastal areas were
earlier interpreted as a megathermal paratropical rain forest
growing under a MAT of ~21-25°C. The new estimate from
CLAMP is 19.4°C (Wolfe, 1994), supporting a paratropical
rain forest-micro thermal mixed broad-leaved evergreen
and coniferous forest. The coolest range would be 10-20°C
(average 15°C). On the basis of LMA, the slightly older
Kupreanof Island flora could represent a notophyllous
broad-leaved evergreen forest or a paratropical rain forest.
The difficulty in interpreting these floras is partly be-
cause the Yakutat block is one of several suspect terranes or
microplates separated from the Wrangell Mountains by the
Denali Fault (Fig. 2.13) and transported from the south. The
extent of the transport is uncertain. Some estimates place
the terrane -20° latitude further south during the Eocene
(Bruns, 1983; Keller et al., 1984). Axelrod (1984, see also
Axelrod et al., 1991) believes Paleogene climates in Alaska
were temperate, and that strata containing tropical ele-
ments from the coastal side of the Denali Fault were carried
up to 1500 km from the south. McKenna notes that

Southwestern Alaska has been the site of tectonic unrest
throughout the Cenozoic as various terranes have ar-
rived from the south. Essentially all of the terranes
south of the Denali Fault were formed at sites many de-
grees south of their present position and have been
mashed against the North American Plate as the Kula
PlaTe and part of the Pacific Plate moved north to de-
struction under North America and eastern Asia [Fig.
2.16]. Thus, environmental information derived from
Paleogene and Mesozoic floras south of the Denali fault
system tells us not about ancient Alaska but about
somewhere else. (1983, p. 469)
Certainly an interpretation of extensive and widespread
tropical floras in the Arctic based on these enigmatic
coastal Paleogene communities (pi 65°-70° N) is not con-
sistent with general global symmetry of vegetation at the
time or with the more meager tropical element found in
other high northern latitude floras (e.g., Musophyllum in
the Thyra 0 flora of Greenland). The adjacent Eocene floras
of northern Japan and Kamschatka are more temperate.
However, other evidence suggests the Yakutat block
may have moved only 5° northward from about the latitude
of British Columbia (Plafker, 1984; Wolfe, 1985; Wolfe and
McCoy, 1984). J. A. Wolfe (personal communication, 1996)
notes that to have moved 20°, the Yukutat block somehow
would have had to jump a triple plate junction. The cur-
rent uncertainty is whether movement was from the vicin-
ity of British Columbia or from as far south as Oregon or
beyond (D. Stone, University of Alaska, personal commu-
nication, 1994). The question is still open (Tiffney, 1985;
see also Kerr, 1995). However, there is no paleomagnetic
evidence to suggest extensive transport during the Paleo-
gene from the western Pacific, so coastal areas can be inter-
preted as supporting tropical vegetation with an Old World
component as least as far north as 45° (Oregon) or 57°
(British Columbia). Northward of 70° N and probably fur-
ther south in the continental interior the vegetation was a
polar broad-leaved deciduous forest.
These floras indicate that during the Paleogene Beringia
was occupied by a polar broad-leaved deciduous forest
with a narrow southern fringe of evergreen megathermal
plants (Tiffney, 1985), although the latter were transported
at least some distance from the south. This tropical flora
with a prominent Old World component is one basis for
the boreotropical flora hypothesis, in contrast to the
Arcto-Tertiary geoflora concept (Chapter 3).
VEGETATION SUMMARY
At the end of the Cretaceous the vegetation of North Amer-
ica consisted of four principal plant formations (Table 5.5).
A tropical forest occupied southeastern United States and
some version extended northward toward Greenland; there
is little data on Maastrichtian vegetation from the interven-
ing region. Genera common to the southern and northern
floras include the form genera Chondrophyllum, Dermato-
Late Cretaceous through Early Eocene History 175
phyllites, and Proteoides, and the extant genera Androm-
eda, Diospyros, and Magnolia. It was an open-canopy for-
est with relatively tall trees with poorly developed growth
rings, and it grew under subhumid conditions. The emer-
gent stratum consisted mostly of early representatives of
the Taxodiaceae (evergreen forms similar to Sequoia; de-
ciduous ones similar to Metasequoia, and the extinct
Parataxodium) and the Cupressaceae (similar to Callitris
and Neocallitropis}. Understory vegetation included ro-
sette palms, Zingiberales (Spirematospermum, Zingiberop-
sis), Magnoliidae (Chloranthaceae), Plantaceae, Fagaceae
(Fagopsis), Juglandaceae, Aceraceae, Ranunculidae, and
Dilleniidae (Theaceae, Euphorbiales, Malvales).
A paratropical forest grew in the southern and middle
regions of the western United States. It was a broad-leaved
evergreen community of subhumid habitats but more
moist than in the southeast. In composition it was similar
to the tropical forest.
The notophyllous broad-leaved evergreen forest occu-
pied areas north of the paratropical forest, and it is rec-
ognized by leaves generally smaller than those in the
evergreen community to the south. Members included
Equisetum, lycopods, Cycadophyte and Ginkgophyte hold-
overs from earlier Cretaceous vegetation, the Araucari-
aceae, evergreen Taxodiaceae, Nyssidium-Joffrea (related
to Cercidiphylluni), Nordenskioldia (related to Trocho-
dendron-Tetracentron), pollen similar to Gunnera, and
Drumhellera of the Taxodium-Metasequoia-Sequoia-
Sequoiadendron complex. Broad-leaved deciduous species
grew along streams and in other disturbed habitats.
A polar broad-leaved deciduous forest occupied the
higher latitudes in North America during the Maastricht-
ian. In the lowlands and swamps, forests of deciduous Tax-
odiaceae were dominant; in better drained areas there was
Ginkgo (along the southern transition margin), trochoden-
droids, Viburnum-type plants, platanoids and other Ham-
amelididae, Betulaceae—Myricaceae, and Ulmaceae.
By the end of the Cretaceous declining temperatures
from mid-Cretaceous highs and drainage of the epiconti-
nental sea (greater seasonality, less poleward transfer of
heat to the interior) and polar light regimes were favoring
diversification and expansion of deciduous microthermal
vegetation in the high latitudes, in disturbed habitats, and
in the few moderately high-altitude sites that existed at the
time. The bolide impact may have intensified this trend re-
gionally.
Although difficult to quantify, four refinements are
emerging in the interpretation of Maastrichtian and Paleo-
gene floras of the high northern latitudes. There is recogni-
tion that tropical vegetation from coastal sites did not
extend widely over inland areas. This more limited em-
phasis on the tropical elements has reduced the degree of
implied random ecological heterogeneity of Paleogene
vegetation. It also indicates that climates and physical con-
ditions were more diverse and less uniformly equable than
earlier thought, allowing the arrangement of plants into
176 Late Cretaceous and Cenozoic History of North American Vegetation

Table 5.5. Summary of North American vegetation types and estimated MATs: Late Cretaceous through Early Eocene.
Names for plant communities follow terminology discussed under Vegetation.

Vegetation
Region Late Cretaceous MAT
Southeast (30°-?° N) Tropical forest 20°C (25° warm intervals)
Northeast (65°-75° N) ? ?10°C
West (40°-46° N) Paratropical 21-22°C
Northwest (50°-5 8° N) Notophyllous broad-leaved evergreen <20°C
North (58°-66° N) Polar broad-leaved deciduous 15°C
North (65°-75° N) Polar broad-leaved deciduous 8°C

K-T Boundary Change

West - northwest Notophyllous broad- 20°C
leaved evergreen to 15°C
polar broad-leaved deciduous

Paleocene
Southeast Tropical rain forest 27°C
Northeast Notophyllous broad-leaved evergreen to 15-20°C
polar broad-leaved deciduous; coastal
megathermal communities
West (High Plains)
Ft. Union, Ecotonal and trending (N-S and 15°C
Sentinel Butte, Early to Late Paleocene)
Bear Den to from polar broad-leaved 13-15/16°C
deciduous forest to
Camels Butte paratropical rain forest-notophyllous 17.8°C
broad-leaved evergreen forest
Powder River Basin Notophyllous broad-leaved evergreen forest 16-18°C
Northwest Trending Early to Late Paleocene
Joffre Bridge (Alberta) to Polar broad-leaved deciduous to notophyllous 10-15°C (from ~8°C in Late
Donnet Plume (Yukon Territory) broad-leaved evergreen Cretaceous)

Early Eocene
Southeast Tropical to subhumid paratropical rain forest >27°C
Northeast ?Notophyllous broad-leaved evergreen (coastal) Avg. ?18°C
to restricted polar broad-leaved deciduous
(inland, uplands)
Plains, eastern Rocky Mts. Subhumid paratropical rain forest ?25°C
West to northwest Tropical to paratropical rain forest (coastal); ?25°C
notophyllous broad-leaved evergreen
to polar broad-leaved deciduous (70°-75°, ?12-15° C
N interior and highlands); or more extensive
paratropical rain forest-broad-leaved
evergreen (70°-75° N) forest (Kushtaka
flora, suspect terrane?) ?21-25°C

habitats and communities more comparable to modern as-
sociations. Also, evidence from plant microfossils reveals a
greater diversity in forest and understory vegetation than
originally suggested by megafossil evidence alone. In terms
of lineages, an emerging generalization is that many Pa-
leocene through Middle Eocene angiosperms, assigned to
modern taxa in the older literature, represent extinct or in-
termediate forms; from the Middle Eocene onward an in-
creasing number can be placed in modern genera based on
details of leaf architecture and the study of associated mul-
tiple organs.
In the earliest Paleocene there was an increase in rain-
fall, which had its most notable effect on the subhumid
vegetation of the southeastern United States. With greater

moisture, the tropical forest was replaced by a more
closed-canopy, multistratal tropical rain forest. Subsequent
climatic changes have left remnants or reintroductions in
southern peninsula Florida and along the coasts of the
southeastern United States (e.g., Rhizophord) and as ele-
ments of tropical origin that evolved temperate descen-
dents (e.g., Diospyros, the only temperate representative of
the primarily tropical family Ebenaceae). The Appalachian
highland provided habitats for more temperate assem-
blages (Alnus, Betula, Carya, and others). Toward the
northeast the vegetation was a notophyllous broad-leaved
evergreen forest, especially along the coast of the North At-
lantic land bridge, grading into a polar broad-leaved decid-
uous forest further inland and in the uplands.
The moist lowlands of the Plains area during the Paleo-
cene and Early Eocene were occupied by a paratropical
rain forest-notophyllous broad-leaved evergreen forest,
grading northward into a notophyllous broad-leaved de-
ciduous forest.
The warming trend continued into the Early Eocene
and culminated with the highest MAT in all of Cenozoic
time. The tropical rain forest persisted in the southeastern
United States but was trending toward a subhumid para-
tropical rain forest near the Middle Eocene. Few paleo-
botanical data are available for the region to the northeast,
but fossil faunas and some evidence from the Eureka
Sound Formation suggest that a paratropical rain forest-
notophyllous broad-leaved evergreen forest likely grew in
the area. The polar broad-leaved deciduous forest, if pres-
ent, was restricted to uplands in the higher latitudes. Para-
tropical rain forest extended into the Plains area and along
the eastern slopes of the Rocky Mountains. On the western
side of the mountains, midlatitude coastal areas were oc-
cupied by a tropical to paratropical rain forest. Interpreta-
tion of the vegetation toward the northwest depends on
the resolution of the Alaskan terrane problem. The longi-
tudinal arrangement of vegetation was from notophyllous
broad-leaved evergreen (coastal) to polar broad-leaved de-
ciduous forest (interior) at 70°-75° N if there was exten-
sive transport of terranes. If there was less transport, the
cline was from paratropical rain forest to notophyllous
broad-leaved evergreen forest with more restricted decidu-
ous forest.
Thus, at the end of the Early Eocene the principal plant
communities of North America were tropical rain forest,
paratropical rain forest, notophyllous broad-leaved ever-
green forest, and a restricted polar broad-leaved decidu-
ous forest. Tropical, subtropical, and warm-temperate veg-
etation had reached the greatest geographic extent it
would achieve; subsequent events favored expansion of
microthermal deciduous communities. As yet there is no
extensive tundra formation, coniferous forest formation
(boreal or montane associations: Appalachian or western
montane coniferous forests), grassland formation, shrub-
land/chaparral-woodland-savanna formation, or desert
formation.
Late Cretaceous through Early Eocene History 177
References
Andrews, H. N., Jr. 1980. The Fossil Hunters, in Search of
Ancient Plants. Cornell University Press, Ithaca, NY.
Areces-Mallea, A. E. 1990. Basopollis krutzchi Kedves:
primera determinacion de un normapolles en el
Paleogeno de Cuba. Cienc. Tierra Espacio 17: 27-32.
Askin, R. A. 1988. The palynological record across the Cre-
taceous/Tertiary transition on Seymour Island, Antarc-
tica. In: R. M. Feldmann and M. O. Woodburne (eds.),
Geology and Paleontology of Seymour Island, Antarc-
tic Peninsula. Geol. Soc. Amer. Mem. 169: 155-162.
Axelrod, D. 1.1952. A theory of angiosperm evolution. Evo-
lution 6: 29-60.
. 1970. Mesozoic paleogeography and early an-
giosperm history. Bot. Rev. 36: 277-319.
. 1984. An interpretation of Cretaceous and Tertiary
biota in polar regions. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 45: 105-147.
. 1992a. What is an equable climate? Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 91:1-12.
. 1992b. Climatic pulses, a major factor in legume
evolution. In: P. S. Herendeen and D. L. Dilcher (eds.),
Advances in Legume Systematics: Part 4. The Fossil
Record. Royal Botanic Gardens, Kew, U.K. pp.
259-279.
-, M. T. K. Arroyo, and P. H. Raven. 1991. Historical
development of temperate vegetation in the Americas.
Rev. Chil. Hist. Nat. 64: 413-446.
Barrick, R. E. and W. J. Showers. 1994. Thermophysiology
of Tyrannosaurus rex: evidence from oxygen isotopes.
Science 265: 222-224.
Barron, E. J. 1983. A warm, equable Cretaceous: the nature
of the problem. Earth-Sci. Rev. 19: 305-338.
. 1984. Climatic implications of the variable obliq-
uity explanation of Cretaceous-Paleogene high-lati-
tude floras. Geology 12: 595-598.
-, J. L. Sloan II, and C. G. A. Harrison. 1980. Potential
significance of land-sea distribution and surface
albedo variations as a climatic forcing factor; 180 M.Y.
to the present. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 30: 17-40.
Barron, E. J., S. L. Thompson, and S. H. Schneider. 1981.
An ice-free Cretaceous? Results from climate model
simulations. Science 212: 501-508.
Barron, E. J., and W. M. Washington. 1982a. Cretaceous cli-
mate: a comparison of atmospheric simulations with
the geologic record. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 40: 103-133.
. 1982b. Atmospheric circulation during warm geo-
logic periods: is the equator-to-pole surface-tempera-
ture gradient the controlling factor? Geology 10:
633-636.
Basinger, J. E, D. R. Greenwood, and T. Sweda. 1994. Early
Tertiary vegetation of Arctic Canada and its relevance
to paleoclimatic interpretation. In: M. C. Boulter and
H. C. Fisher (eds.), Cenozoic Plants and Climates of the
Arctic. NATO ASI Series 1: Global Environmental
Change. Springer-Verlag, Berlin. Vol. 27, pp. 175-198.
Batten, D. J. 1984. Palynology, climate and the development
of Late Cretaceous floral provinces in the northern
hemisphere: a review. In: P. Brenchley (ed.), Fossils
and Climate. Wiley, New York. pp. 127-164.
Beeson, D. C. 1992. High resolution palynostratigraphy
178 Late Cretaceous and Cenozoic History of North American Vegetation
across a marine Cretaceous-Tertiary boundary inter-
val, Falls County, Texas. Am. Assoc. Stratig. Palynol.
Newslett. 25:17-18.
Bell, W. A. 1949. Uppermost Cretaceous and Paleocene flo-
ras of western Alberta. Geol. Surv. Can. Bull. 13:1-231.
Berggren, W. A. 1982. Role of ocean gateways in climatic
change. In: Geophysics Study Committee (eds.), Cli-
mate in Earth History. National Academy Press, Wash-
ington, B.C. pp. 118-125.
, D. V Kent, M.-P. Aubry, and J. Hardenbol (eds.).
1995. Geochronology, Time Scales and Global Strati-
graphic Correlation. Society of Sedimentary Geology,
Tulsa, OK. Special Publ. 54.
Berner, R. A., A. C. Lasaga, and R. M. Garrels. 1983. The
carbonate-silicate geochemical cycle and its effect on
atmospheric carbon dioxide over the past 100 million
years. Am. J. Sci. 283: 641-683.
Berry, E. W. 1916. The Lower Eocene Floras of Southeast-
ern North America. U.S. Geological Survey, Reston,
VA. Professional Paper 91, pp. 1-481.
. 1922. Additions to the flora of the Wilcox Group.
U.S. Geological Survey, Reston, VA. Professional Paper
131-A, pp. 1-21.
. 1924. The Middle and Upper Eocene Floras of
Southeastern North America. U.S. Geological Survey,
Reston, VA. Professional Paper 92, pp. 1-206.
. 1925. The Flora of the Ripley Formation. U.S. Geo-
logical Survey, Reston, VA. Professional Paper 136, pp.
1-94.
. 1930. Revision of the Lower Eocene Wilcox Flora of
the Southeastern States. U.S. Geological Survey, Re-
ston, VA. Professional Paper 156, pp. 1-196.
. 1935. A preliminary contribution to the floras of
the Whitemud and Ravenscrag formations. Geol. Surv.
Can. Mem. 182: 1-107.
-. 1941. Additions to the Wilcox Flora from Kentucky
and Texas. U.S. Geological Survey, Reston, VA. Profes-
sional Paper 193-E, pp. 83-99.
Bown, T. M., K. D. Rose, E. L. Simons, and S. L. Wing. 1994.
Distribution and Stratigraphic Correlation of Upper
Paleocene and Lower Eocene Fossil Mammal and Plant
Localities of the Fort Union, Willwood, and Tatman
Formations, Southern Big Horn Basin, Wyoming. U.S.
Geological Survey, Reston, VA. Professional Paper
1540, pp. 1-103.
Boyd, A. 1990. The Thyra 0 flora: toward an understand-
ing of the climate and vegetation during the Early Ter-
tiary in the high Arctic. Rev. Palaeobot. Palynol. 62:
189-203.
. 1992. Musopsis N. Gen.: a banana-like leaf genus
from the Early Tertiary of eastern north Greenland.
Am. J. Bot. 79: 1359-1367.
Brouwers, E. M. (+ 5 authors). 1987. Dinosaurs on the
North Slope, Alaska: high latitude, latest Cretaceous
environments. Science 237: 1608-1610.
Brown, R. W. 1962. Paleocene flora of the Rocky Mountains
and Great Plains. U.S. Geological Survey, Reston, VA.
Professional Paper 375, pp. 1-119.
Bruns, T. R. 1983. Model for the origin of the Yakutat block,
an accreting terrane in the northern Gulf of Alaska. Ge-
ology 11: 718-721.
Call, V. B. and D. L. Dilcher, 1997. The fossil record of Eu-
commia (Eucommiaceae) in North America. Am. J.
Bot. 84: 798-814.
Chandrasekharam, A. 1974. Megafossil flora from the
Genesee locality, Alberta, Canada. Palaeontograph.,
Abt. B, 147: 1-41.
Clemens, W. A. and C. W. Allison. 1985. Late Cretaceous
terrestrial vertebrate fauna, North Slope, Alaska. Geol.
Soc. Am. Abstr. Programs 17: 548.
and L. G. Nelms. 1993. Paleoecological implications
of Alaskan terrestrial vertebrate fauna in latest Creta-
ceous time at high paleolatitudes. Geology 21:
503-506.
Crane, R R. 1987. Vegetational consequences of the an-
giosperm diversification. In: E. M. Friis, W. G. Chaloner,
and R R. Crane (eds.), The Origins of Angiosperms and
Their Biological Consequences. Cambridge University
Press, Cambridge, U.K. pp. 107-144.
. 1989. Paleobotanical evidence on the early radia-
tion of nonmagnoliid dicotyledons. Plant Syst. Evol.
162: 165-191.
-, S. R. Manchester, and D. L. Dilcher. 1990. A prelim-
inary survey of fossil leaves and well-preserved repro-
ductive structures from the Sentinel Butte Formation
(Paleocene) near Almont, North Dakota. Fieldiana (Ge-
ology) 20:1-63.
Crane, R R. and R. A. Stockey. 1985. Growth and reproduc-
tive biology of Joffrea speirsii gen. et sp. nov., a Cer-
didiphyllum-like plant from the Late Paleocene of Al-
berta, Canada. Can. J. Bot. 63: 340-364.
Creber, G. T. and W. G. Chaloner. 1985. Tree growth in the
Mesozoic and Early Tertiary and the reconstruction of
palaeoclimates. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 52: 35-60.
Crepet, W. L. 1981. The status of certain families of the
Amentiferae during the Middle Eocene and some hy-
potheses regarding the evolution of wind pollination
in dicotyledonous angiosperms. In: K. J. Niklas (ed.),
Paleobotany, Paleoecology, and Evolution. Praeger,
New York. Vol. 2, pp. 103-128.
. 1989. History and implications of the early North
American fossil record of Fagaceae. In: R R. Crane and
S. Blackmore (eds.), Evolution, Systematics, and Fossil
History of the Hamamelidae, Vol. 2, "Higher"
Hamamelidae. Clarendon Press, Oxford, U.K. Special
Vol. 40B, pp. 45-66.
and C. P. Daghlian. 1981. Lower Eocene and Paleo-
cene Gentianaceae: floral and palynological evidence.
Science 214: 75-77.
Crepet, W. L. and G. D. Feldman. 1991. The earliest remains
of grasses in the fossil record. Am. J. Bot. 78: 1010-
1014.
Crepet, W. L. and R S. Herendeen. 1992. Papilionoid flow-
ers from the Early Eocene of southeastern North Amer-
ica. In: R S. Herendeen and D. L. Dilcher (eds.), Ad-
vances in Legume Systematics: Part 4. The Fossil
Record. Royal Botanic Gardens, Kew, U.K. pp. 43-55.
Crepet, W. L. and K. C. Nixon. 1989a. Earliest megafossil
evidence of Fagaceae: phylogenetic and biogeographic
implications. Am. J. Bot. 76: 842-855.
. 1989b. Extinct transitional Fagaceae from the
Oligocene and their phylogenetic implications. Am. J.
Bot. 76: 1493-1505.
Crepet, W. L. and D. W. Taylor. 1985. The diversification of
the Leguminosae: first fossil evidence of the Mi-
mosoideae and Papilionoideae. Science 228: 1087-
1089.

. 1986. Primitive mimosoid flowers from the Paleo-
cene-Eocene and their systematic and evolutionary
implications. Am. J. Bot. 73: 548-563.
Crowley, T. J. and G. R. North. 1991. Paleoclimatology. Ox-
ford Monographs in Geology and Geophysics. Oxford
Univ. Press, Oxford, U.K. No. 16, Chap. 9.
Crowley, T. J., D. A. Short, J. G. Mengel, and G. R. North.
1986. Role of seasonality in the evolution of climate
during the past 100 million years. Science 231:
579-584.
Dawson, M. R., L. Krishtalka, and R. K. Stucky. 1990. Revi-
sion of the Wind River faunas, Early Eocene of central
Wyoming. Part 9. The oldest known Hystricomor-
phous rodent (Mammalia:Rodentia). Ann. Carnegie
Mus. 59: 135-147.
Dawson, M. R., R. M. West, W. Langston, Jr., and }. H.
Hutchison. 1976. Paleogene terrestrial vertebrates:
northernmost occurrence, Ellesmere Island, Canada.
Science 192: 781-782.
Dilcher, D. L. 1971. A revision of the Eocene flora of south-
eastern North America. Palaeobotanist 20: 7-18.
. 1973. A paleoclimatic interpretation of the Eocene
floras of southeastern North Amerca. In: A. Graham
(ed.), Vegetation and Vegetational History of Northern
Latin America. Elsevier Science Publishers, Amster-
dam, pp. 39-59.
Dolph, G. E. and D. L. Dilcher. 1979. Foliar physiognomy as
an aid in determining paleoclimate. Palaeontograph.,
Abt. B, 170: 151-172.
Dorf, E. 1942. Upper Cretaceous floras of the Rocky Moun-
tain region. Carnegie Inst. Wash. Contrib. Paleontol.
508.
. 1960. Tertiary fossil forests of Yellowstone National
Park, Wyoming. Billings Geological Society of Amer-
ica Guidebook, Geological Society of America, Denver,
CO, llth Annual Field Conference, pp. 253-260.
Elsik, W. C. 1968a. Palynology of a Paleocene Rockdale lig-
nite, Milam County, Texas. I. Morphology and Taxon-
omy. Pollen Spores 10: 263-314.
. 1968b. Palynology of a Paleocene Rockdale lignite,
Milan County, Texas. II. Morphology and Taxonomy
(end). Pollen Spores 10: 599-664.
-. 1970. Palynology of a Paleocene Rockdale lignite,
Milam County, Texas. III. Errata and taxonomic revi-
sions. Pollen Spores 12: 99-101.
Emanuel, K. A., K. Speer, R. Rotunno, R. Srivastava, and M.
Molina. 1995. Hypercanes: a possible link in global ex-
tinction scenarios. J. Geophys. Res. 100D:
13755-13765.
Estes, R. and J. H. Hutchison. 1980. Eocene lower verte-
brates from Ellesmere Island, Canadian Arctic Archi-
pelago. Palaeogeogr., Palaeoclimatol., Palaeoecol. 30:
325-347.
Farabee, M. J. 1990. Triprojectate fossil pollen genera. Rev.
Palaeobot. Palynol. 65: 341-347.
. 1991. Botanical affinities of some Triprojectacites
fossil pollen. Am. J. Bot. 78: 1172-1181.
-. 1993. Morphology of triprojectate fossil pollen:
form and distribution in space and time. Bot. Rev. 59:
211-249.
Farley, M. B. 1990. Vegetation distribution across the early
Eocene depositional landscape from palynological
analysis. Palaeogeogr., Palaeoclimatol., Palaeoecol. 79:
11-27.
Late Cretaceous through Early Eocene History 179
Ferguson, K. M., T. M. Lehman, and R. T. Gregory. 1991. C-
and O-isotopes of pedogenic soil nodules from two
sections spanning the K/T transition in west Texas.
Geol. Soc. Am., Abstr. Program 23: 302.
Fleming, R. E and D. J. Nichols. 1990. The fern-spore abun-
dance anomaly at the Cretaceous-Tertiary boundary: a
regional bioevent in western North America. In: E. G.
Kauffman and O. H. Walliser (eds.), Extinction Events
in Earth History. Springer-Verlag, Berlin, pp. 347-350.
Foster, H. L. and Y. Igarashi. 1990. Fossil pollen from non-
marine sedimentary rocks of the eastern Yukon-
Tanana region, east-central Alaska. U.S. Geol. Surv.
Bull. 1946: 11-20.
Frakes, L. A. 1979. Climates Throughout Geologic Time. El-
sevier Science Publishers, Amsterdam.
Francis, J. E. 1988. A 50 million-year old forest from Strath-
cona Fiord, Ellesmere Island, Arctic Canada: evidence
for a warm polar climate. Arctic 41: 314-318.
Franczyk, K. J., T. D. Fouch, R. C. Johnson, C. M. Molenaar,
and W. A. Cobban. 1992. Cretaceous and Tertiary
paleogeographic reconstructions for the
Uinta—Piceance Basin study area, Colorado and Utah.
U.S. Geol. Surv. Bull. 1787-Q: 1-37.
Frederiksen, N. O. 1987. Tectonic and paleogeographic set-
ting of a new latest Cretaceous floristic province in
North America. Palaios 2: 533-542.
. 1988. Sporomorph Biostratigraphy, Floral Changes,
and Paleoclimatology, Eocene and Earliest Oligocene
of the Eastern Gulf Coast. U.S. Geological Survey,
Reston, VA. Professional Paper 1448.
. 1989a. Changes in floral diversities, floral turnover
rates, and climates in Campanian and Maastrichtian
time, North Slope of Alaska. Cretaceous Res. 10:
249-266.
. 1989b. Late Cretaceous and Tertiary floras, vegeta-
tion, and paleoclimates of New England. Rhodora 91:
25-48.
. 1991a. Rates of floral turnover and diversity change
in the fossil record. Palaeobotanist 39:127-139.
. 1991b. Pulses of Middle Eocene to earliest
Oligocene climate deterioration in southern California
and the Gulf Coast. Palaios 6: 564-571.
. 1994. Paleocene floral diversities and turnover
events in eastern North America and their relation to
diversity models. Rev. Palaeobot. Palynol. 82:
225-238.
, T. A. Ager, and L. E. Edwards. 1988. Palynology of
Maastrichtian and Paleocene rocks, lower Colville
River region, North Slope of Alaska. Can. J. Earth Sci.
25: 512-527.
Friis, E. M. 1983. Upper Cretaceous (Senonian) floral struc-
tures of juglandalean affinity containing normapolles
pollen. Rev. Palaeobot. Palynol. 39: 161-188.
. 1987. Spriematospermum chandlerae sp. nov., an
extinct species of Zingiberaceae from the North Amer-
ican Cretaceous. Tert. Res. 9: 7-12.
Greenwood, D. R. and J. F Basinger. 1994. The paleoecology
of high-latitude Eocene swamp forests from Axel
Heiberg Island, Canadian high Arctic. Rev. Palaeobot.
Palynol. 81: 83-97.
Greenwood, D. R. and S. L. Wing. 1995. Eocene continental
climates and latitudinal temperature gradients. Geol-
ogy 23: 1044-1048.
Habicht, J. K. A. 1979. Paleoclimate, paleomagnetism, and
180 Late Cretaceous and Cenozoic History of North American Vegetation
continental drift. In: American Association Petroleum
Geology (eds.), Studies in Geology 9. American Asso-
ciation of Petroleum Geology, Tulsa, OK.
Hallam, A. 1984. Continental humid and arid zones during
the Jurassic and Cretaceous. Palaeogeogr., Palaeoclima-
tol., Palaeoecol. 47: 195-223.
Harland, W. B. (+ 5 authors). 1990. A Geologic Time Scale
1989. Cambridge University Press, Cambridge, U.K.
Heer, O. 1883. Den Zweiten Theil der fossilen Flora Gron-
lands. Flora Fossilis Arctica. J. Wurster, Zurich. Vol. 7.
Herendeen, P. S., D. H. Les, and D. L. Dilcher. 1990. Fossil
Ceratophyllum (Ceratophyllaceae) from the Tertiary of
North America. Am. J. Bot. 77: 7-16.
Herman, A. B. 1994. Late Cretaceous Arctic platanoids and
high latitude climate. In: M. C. Boulter and H. C.
Fisher (eds.), Cenozoic Plants and Climates of the Arc-
tic. NATO ASI Series 1: Global Environmental Change.
Springer-Verlag, Berlin. Vol. 27, pp. 151-159.
and R. A. Spicer. 1996. Palaeobotanical evidence for
a warm Cretaceous Arctic Ocean. Nature 380: 330-333.
-. 1997. New quantitative palaeoclimate data for the
Late Cretaceous Arctic: evidence for a warm polar
ocean. Palaeogeogr., Palaeoclimatol., Palaeoecol. 128:
227-251.
Herngreen, G. E W. and A. E Chlonova. 1981. Cretaceous
microfloral provinces. Pollen Spores 23: 441-555.
Rickey, L. J. 1977. Stratigraphy and paleobotany of the
Golden Valley Formation (Early Tertiary) of western
North Dakota. Geol. Soc. Am. Mem. 150: 1-183.
. 1980. Paleocene stratigraphy and flora of the Clark's
Fork Basin. In: P. D. Gingerich (ed.), Early Cenozoic Pa-
leontology and Stratigraphy of the Bighorn Basin,
Wyoming. Univ. Mich. Papers Paleontol. 24: 33-49.
. 1981. Land plant evidence compatible with grad-
ual, not catastrophic, change at the end of the Creta-
ceous. Nature 292: 529-531.
-. 1984. Changes in the angiosperm flora across the
Cretaceous-Tertiary boundary. In: W. A. Berggren and
J. A. Van Couvering (eds.), Catastrophes and Earth His-
tory, the New Uniformitarianism. Princeton University
Press, Princeton, NJ. pp. 279-337.
Hickey, L. J. and L. J. McWeeney. 1992. Plants at the K/T
boundary. Nature (Sci. Corresp.) 356: 295-296.
Hickey, L. J. and R. K. Peterson. 1978. Zingiberopsis, a fos-
sil genus of the ginger family from Late Cretaceous to
Early Eocene sediments of Western Interior North
America. Can. J. Bot. 56: 1136-1152.
Hickey, L. J., R. M. West, M. R. Dawson, and D. K. Choi.
1983. Arctic terrestrial biota: paleomagnetic evidence
of age disparity with mid-northern latitudes during
the Late Cretaceous and Early Tertiary. Science 221:
1153-1156.
Hoffman, A. 1989. Changing palaeontological views on
mass extinction phenomena. In: S. K. Donovan (ed.),
Mass Extinctions, Processes and Evidence. Columbia
University Press, New York. pp. 1-18.
Hoffman, G. L. and R. A. Stockey. 1992. Vegetative and fer-
tile remains of Azolla from the Paleocene Joffre Bridge
locality, Alberta. Am. J. Bot. 79(Suppl.): 102-103.
Horrell, M. A. 1991. Phytogeography and paleoclimatic in-
terpretation of the Maestrichtian. Palaeogeogr., Palaeo-
climatol., Palaeoecol. 86: 87-138.
Hsu, K. J. and J. A. McKenzie. 1985. A "Strangelove" ocean
in the earliest Tertiary. In: E. T. Sundquist and W. S.
Broecker (eds.), The Carbon Cycle and Atmospheric
CO2: Natural Variations Archean to Present. Geophys-
ical Monograph 32. American Geophysical Union,
Washington, D.C. pp. 487-492.
Hutchison, J. H. 1982. Turtle, crocodilian, and champ-
sosaur diversity changes in the Cenozoic of the north-
central region of western United States. Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 37:149-164.
Jarzen, D. M. 1977. Aquilapollenites and some Santalalean
genera. Grana 16: 29-39.
. 1980. The occurrence of Gunnera pollen in the fos-
sil record. Biotropica 12: 117-123.
. 1982a. Angiosperm pollen from the Ravenscrag
Formation (Paleocene), southern Saskatchewan,
Canada. Pollen Spores 14:119-155.
. 1982b. Palynology of Dinosaur Provincial Park
(Campanian) Alberta. Syllogeus 38: 1-69.
and M. E. Dettmann. 1989. Taxonomic revision of
Tricolpites reticulatus Cookson ex Couper, 1953 with
notes on the biogeography of Gunnera L. Pollen
Spores 31: 97-112.
Jerzykiewicz, T. and A. R. Sweet. 1986. The Cretaceous-
Tertiary boundary in the central Alberta foothills. I.
Stratigraphy. Can. J. Earth Sci. 23:1356-1374.
Johnson, K. R. 1992. Leaf-Fossil evidence for extensive flo-
ral extinction at the Cretaceous-Tertiary boundary,
North Dakota, USA. Cretaceous Res. 13: 91-117.
. 1993. Extinctions at the Antipodes. Nature 366:
511-512.
-. 1996. Description of seven common leaf species
from the Hell Creek Formation (Upper Cretaceous:
Upper Maastrichtian), North Dakota, South Dakota,
and Montana. Proc. Denver Museum Natural History,
Ser. 3, No. 12: 1-47.
Johnson, K. R. and L. J. Hickey. 1990. Megafloral Change
Across the Cretaceous/Tertiary Boundary in the North-
ern Great Plains and Rocky Mountains, U.S.A. Geolog-
ical Society of America, Boulder, CO. Special Paper
247, pp. 433-444.
Johnson, K. R., D. J. Nichols, M. Attrep, Jr., and C. J. Orth.
1989. High-resolution leaf-fossil record spanning the
Cretaceous/Tertiary boundary. Nature 340: 708-711.
Jones, J. H. and D. L. Dilcher. 1988. A study of the
"Dryophyllum" leaf forms from the Paleogene of
southeastern North America. Palaeontograph., Abt. B,
208: 53-80.
Kalgutkar, R. M. and D. J. Mclntyre. 1991. Helicosporous
fungi and Early Eocene pollen, Eureka Sound Group,
Axel Heiberg Island, Northwest Territories. Can. J.
Earth Sci. 28: 364-371.
Kalkreuth, W. D., D. J. Mclntyre, and R. J. H. Richardson.
1993. The geology, petrology and palynology of Ter-
tiary coals from the Eureka Sound Group at Strathcona
Fiord and Bache Peninsula, Ellesmere Island, Arctic
Canada. Int. J. Coal Geol. 24: 75-111.
Kauffman, E. G. 1977. Geological and biological overview:
western interior Cretaceous basin. Mt. Geol. 14:
75-100.
Kay, M. Geologic Map of North America. Wiley, New York.
[Distributed by the Geological Society of America, no
scale.]
Keller, G. 1993. The Cretaceous-Tertiary boundary transi-
tion in the Antarctic Ocean and its global implica-
tions. Mar. Micropaleontol. 21: 1-45.

, E. Barrera, B. Schmitz, and E. Mattson. 1993. Grad-
ual mass extinction, species survivorship, and long-
term environmental changes across the Cretaceous -
Tertiary boundary in high latitudes. Geol. Soc. Am.
Bull. 105: 979-997.
Keller, G., R. Von Huene, K. McDougall, and T. R. Bruns.
1984. Paleoclimatic evidence for Cenozoic migration
of Alaska terranes. Tectonics 3: 473-495.
Kent, D. V, M. C. McKenna, N. D. Opdyke, J. J. Flynn, and B.
J. MacFadden. 1984. Arctic biostratigraphic hetero-
chroneity. Sci. (Lett): 224:173-174.
Kerr, R. A. 1992. Did an asteroid leave its mark in Montana
bones? Science 256:1395-1396.
Kerr, R. A. 1995. How far did the west wander? Science
268:635-637.
Knowlton, F. H. 1899. Fossil flora of the Yellowstone Na-
tional Park. U.S. Geol. Surv. Monogr. 32: 651-882.
. 1917. Fossil floras of the Vermejo and Raton forma-
tions of Colorado and New Mexico. U.S. Geological
Survey, Reston, VA. Professional Paper 101, pp.
223-455.
. 1922. The Laramie flora of the Denver Basin, with
a review of the Laramie problem. U.S. Geological Sur-
vey, Reston, VA. Professional Paper 130, pp. 1-175.
. 1930. The flora of the Denver and associated forma-
tions of Colorado. U.S. Geological Survey, Reston, VA.
Professional Paper 155, pp. 1-142.
Koch, B. E. 1963. Fossil plants from the lower Paleocene of
the Agatdalen (Angmartussut) area, central Nugssuaq
Peninsula, northwest Greenland. Meddelelser Gr0n-
land 172: 1-120.
. 1964. Review of fossil floras and nonmarine de-
posits of west Greenland. Geol. Soc. Am. Bull. 75:
535-548.
Koch, P. L., J. C. Zachos, and P. D. Gingerich. 1992. Correla-
tion between isotope records in marine and continen-
tal carbon reservoirs near the Palaeocene/Eocene
boundary. Nature 358: 319-322.
Koeberl, C. and R. R. Anderson (eds.). 1996. The Manson
impact structure, Iowa: anatomy of an impact crater.
Geol. Soc. Am. Sp. Paper 302:1-482.
Leffmgwell, H. A. 1971 (1970). Palynology of the Lance
(Late Cretaceous) and Fort Union (Paleocene) forma-
tions of the type Lance area, Wyoming. In: R. M.
Kosanke and A. T. Cross (eds.), Symposium on Paly-
nology of the Late Cretaceous and Early Tertiary. Geo-
logical Society of America, Boulder, CO. Special Paper
127, pp. 1-64.
Lehman, T. M. 1990. Paleosols and Cretaceous/Tertiary
transition in the Big Bend region of Texas. Geology 18:
362-364.
Leopold, E. B. and H. D. MacGinitie. 1972. Development
and affinities of Tertiary floras in the Rocky Moun-
tains. In: A. Graham (ed.), Floristics and Paleofloristics
of Asia and Eastern North America. Elsevier Science
Publishers, Amsterdam, pp. 147-200.
LePage, B. A. and J. F Basinger. 1991a. A new species of
Larix (Pinaceae) from the Early Tertiary of Axel
Heiberg Island, Arctic Canada. Rev. Palaeobot. Palynol.
70:89-111.
. 1991b. Early Tertiary Larix from the Buchanan Lake
Formation, Canadian Arctic Archipelago and a consid-
eration of the phytogeography of the genus. In: R. L.
Christie and N. J. McMillan (eds.), Tertiary Fossil
Late Cretaceous through Early Eocene History 181
Forests of the Geodetic Hills, Axel Heiberg Island, Arc-
tic Archipelago. Geological Survey of Canada Bull.
403: 67-82.
—. 1995a. Evolutionary history of the genus
Pseudolarix Gordon (Pinaceae). Internat. J. Plant Sci.
156: 910-950.
-. 1995b. The evolutionary history of the genus Larix
(Pinaceae). U.S. Dept. Agriculture, Forest Service, Gen.
Tech. Rept. GTR-INT-319: 19-29.
Lidgard, S. and P. R. Crane. 1990. Angiosperm diversifica-
tion and Cretaceous floristic trends: a comparison of
palynofloras and leaf macrofloras. Paleobiology 16:
77-93.
Lloyd, C. R. 1982. The mid-Cretaceous Earth: paleogeogra-
phy; ocean circulation and temperature; atmospheric
circulation. J. Geol. 90: 393-414.
Lowenstam, H. A. and S. Epstein. 1954. Paleotemperatures
of the post-Aptian Cretaceous as determined by the
oxygen isotope method. J. Geol. 62: 207-248.
MacGinitie, H. D. 1941. A Middle Eocene flora from the
central Sierra Nevada. Carnegie Inst. Wash. Contrib.
Paleontol. 534.
Magallon-Puebla, S. and S. Cevallos-Ferriz. 1994. Eucom-
mia constans n. sp. Fruits from upper Cenozoic strata
of Puebla, Mexico: morpholgical and anatomical com-
parison with Eucommia ulmoides Oliver. Internat. J.
Plant Sci. 155: 80-95.
Manchester, S. R. 1987. The fossil history of the Juglan-
daceae. Monogr. Syst. Bot. Missouri Bot. Garden 21:
1-137.
. 1991. Cruciptera, a new juglandaceous winged
fruit from the Eocene and Oligocene of western North
America. Syst. Bot. 16: 715-725.
Manchester, S. R. and Z.-D. Chen. 1996. Palaeocarpinus as-
pinosa sp. Nov. (Betulaceae) from the Paleocene of
Wyoming, U.S.A. Internat. J. Plant Sci. 157: 644-655.
, P R. Crane, and D. L. Dilcher. 1991. Nordenskioldia
and Trochodendron (Trochodendraceae) from the
Miocene of northwestern North America. Bot. Gaz.
152: 357-368.
Manchester, S. R. and D. L. Dilcher. 1982. Pterocaryoid
fruits (Juglandaceae) in the Paleogene of North Amer-
ica and their evolutionary and biogeographic signifi-
cance. Am. J. Bot. 69: 275-286.
Manchester, S. R. and D. L. Dilcher. 1997. Reproductive and
vegetative morphology of Polyptera (Juglandaceae)
from the Paleocene of Wyoming and Montana. Am. J.
Bot. 84: 649-663.
Manchester, S. R. and W. J. Kress. 1993. Fossil bananas
(Musaceae): Ensete oregonense sp. nov. from the
Eocene of western North America and its phytogeo-
graphic significance. Am. J. Bot. 80: 1264-1272.
McGookey, D. P. et al. (+ 6 authors). 1972. Cretaceous sys-
tem. In: W. W. Mallory (ed.), Geologic Atlas of the
Rocky Mountain Region. Rocky Mountain Association
of Geology, Denver, CO. pp. 190-228.
Mclntyre, D. J. 1989. Paleocene palynoflora from northern
Somerset Island, District of Franklin, N.W.T. Current
Res., Geol. Surv. Can. 89-lG: 191-197.
and B. D. Ricketts. 1989. New palynological data
from Cornwall Arch, Cornwall and Amund Ringnes is-
lands, District of Franklin, N.W.T. Current Res., Geol.
Surv. Can. 89-lG: 199-202.
Mclver, E. E. and J. E Basinger. 1993. Flora of the Raven-
182 Late Cretaceous and Cenozoic History of North American Vegetation
scrag Formation (Paleocene), southwestern Saskatche-
wan, Canada. Palaeontograph. Can. 10:1-167.
McKenna, M. C. 1980. Eocene paleolatitude, climate, and
mammals of Ellesmere Island. Palaeogeogr., Palaeocli-
matol., Palaeoecol. 30: 349-362.
. 1983. Holarctic landmass rearrangement, cosmic
events, and Cenozoic terrestrial organisms. Ann. Mis-
souri Bot. Garden 70: 459-489.
Muller, R. A. and G. J. MacDonald. 1997. Glacial cycles and
astronomical forcing. Science 277: 215-218.
Nichols, D. J. 1992. Plants at the K/T boundary. Nature (Sci.
Corresp.): 356: 295.
and R. E Fleming. 1990. Plant microfossil record of
the terminal Cretaceous event in the western United
States and Canada. In: V. L. Sharpton and P. D. Ward
(eds.), Global Catastrophes in Earth History. Geological
Society of America, Boulder, CO. Special Paper 247,
pp. 445-455.
and N. O. Frederiksen. 1990. Palynological evidence
of effects of the terminal Cretaceous event on terres-
trial floras in western North America. In: E. G. Kauff-
man and O. H. Walliser (eds.), Extinction Events in
Earth History. Springer-Verlag, Berlin, pp. 351-364.
Nichols, D. J., D. M. Jarzen, C. J. Orth, and P. Q. Oliver. 1986.
Palynological and iridium anomalies at the Cretaceous-
Tertiary boundary, south-central Saskatchewan. Science
231: 714-717.
Nichols, D. J. and H. L. Ott. 1978. Biostratigraphy and evo-
lution of the Momipites-Caryapollenites lineage in the
Early Tertiary in the Wind River Basin, Wyoming. Pal-
ynology 2: 93-112.
Nichols, D. J., J. A. Wolfe, and D. T. Pocknall. 1989. Latest
Cretaceous and Early Tertiary history of vegetation in
the Powder River Basin, Montana and Wyoming. 28th
Int. Geol. Congr. Guidebook T132: 28-33.
Nixon, K. C. 1989. Origins of Fagaceae. In: P. R. Crane and
S. Blackmore (eds.), Evolution, Systematics, and Fossil
History of the Hamamelidae, Vol. 2, "Higher"
Hamamelidae. Clarendon Press, Oxford, U.K. Special
Vol. 40B, pp. 23-43.
and W. L. Crepet. 1989. Trigonobalanus (Fagaceae):
taxonomic status and phylogenetic relationships. Am.
J. Bot. 76: 828-841.
Norris, G. and A. D. Miall. 1984. Arctic biostratigraphic
heterochroneity. Sci. (Lett.): 224: 174-175.
Norton, N. J. and J. W. Hall. 1967. Guide sporomorphae in
the Upper Cretaceous-Lower Tertiary of eastern Mon-
tana (U.S.A.). Rev. Palaeobot. Palynol. 2: 99-110.
. 1969. Palynology of the Upper Cretaceous and
Lower Tertiary in the type locality of the Hell Creek
Formation, Montana, U.S.A. Palaeontograph., Abt. B,
128: 1-64.
Oltz, D. F, Jr. 1969. Numerical analyses of palynological
data from Cretaceous and Early Tertiary sediments in
east central Montana. Palaeontograph., Abt. B, 128:
90-166.
Opdyke, N. D. 1990. Magnetic stratigraphy of Cenozoic ter-
restrial sediments and mammalian dispersal. J. Geol.
98:621-637.
Orth, C. J., J. S. Gilmore, J. D. Knight, C. L. Pillmore, and
R. H. Tschudy. 1981. An iridium abundance anomaly
at the palynological Cretaceous-Tertiary boundary in
northern New Mexico. Science 214: 1341-1343.
Otto-Bliesner, B. L. and G. R. Upchurch, Jr. 1997. Vegeta-
tion-induced warming of high-latitude regions during
the Late Cretaceous period. Nature 385: 804-807.
Owen, R. M. and D. K. Rea. 1985. Sea-floor hydrothermal
activity links climate to tectonics: the Eocene carbon
dioxide greenhouse. Science 227: 166-169.
Pabst, M. B. 1968. The flora of the Chuckanut Formation of
northwestern Washington, the Equisetales, Filicales,
and Coniferales. Univ. Calif. Publ. Geol. Sci. 76: 1-85.
Page, V. M. 1979. Dicotyledonous wood from the Upper
Cretaceous of central California. J. Arnold Arb. 60:
323-349.
. 1980. Dicotyledonous wood from the Upper Creta-
ceous of central California, II. J. Arnold Arb. 61:
723-748.
. 1981. Dicotyledonous wood from the Upper Creta-
ceous of central California. III. Conclusions. J. Arnold
Arb. 62: 437-455.
Parrish, J. T. 1987. Global palaeogeography and palaeocli-
mate of the Late Cretaceous and Early Tertiary. In:
E. M. Friis, W. G. Chaloner, and R R. Crane (eds.), The
Origins of Angiosperms and Their Biological Conse-
quences. Cambridge University Press, Cambridge, U.K.
pp. 51-73.
and R. A. Spicer. 1988. Late Cretaceous terrestrial
vegetation: a near-polar temperature curve. Geology
16: 22-25.
Pedersen, K. R. 1976. Fossil floras of Greenland. In: A. Es-
cher and W. Swatt (eds.), Geology of Greenland. Geo-
logical Survey of Greenland, Copenhagen, pp.
519-535.
Pigg, K. B. and R. A. Stockey. 1991. Platanaceous plants
from the Paleocene of Alberta, Canada. Rev. Palaeobot.
Palynol. 70: 125-146.
and S. L. Maxwell. 1993. Paleomyrtinaea, a new
genus of permineralized myrtaceous fruits and seeds
from the Eocene of British Columbia and Paleocene of
North Dakota. Can. J. Bot. 71:1-9.
Plafker, G. 1984. Comments and replies on "Model for the
origin of the Yakutat block, an accreting terrane in the
northern Gulf of Alaska." Geology 12: 563.
Pocknall, D. T. 1987. Palynomorph biozones for the Fort
Union and Wasatch formations (Upper Paleocene-
Lower Eocene), Powder River Basin, Wyoming and
Montana, U.S.A. Palynology 11: 23-35.
Ramanujan, C. G. K. and W. N. Stewart. 1969. Fossil woods
of Taxodiaceae from the Edmonton Formation (Upper
Cretaceous) of Alberta. Can. J. Bot. 47: 115-124.
Rea, D. K., M. Leinen, and T. R. Janecek. 1985. Geologic ap-
proach to the long-term history of atmospheric circula-
tion. Science 277: 721-725.
Rea, D. K., J. C. Zachos, R. M. Owen, and R D. Gingerich.
1990. Global change at the Paleocene-Eocene bound-
ary: climatic and evolutionary consequences of tec-
tonic events. Palaeogeogr., Palaeoclimatol., Palaeoecol.
79: 117-128.
Responses. 1988. Science 239: 10-11.
Retallack, G. J. 1994. A peodtype approach to latest Creta-
ceous and earliest Tertiary paleosols in eastern Mon-
tana. Geol. Soc. Am. Bull. 106: 1377-1397.
Richards, P. W. 1952. The Tropical Rain Forest, an Ecologi-
cal Study. Cambridge University Press, Cambridge,
U.K.
Rose, K. D., L. Krishtalka, and R. K. Stucky. 1991. Revision
of the Wind River faunas, Early Eocene of central

Wyoming. Part 11. Palaeanodonta (Mammalia). Ann.
Carnegie Mus. 60: 63-82.
Rothwell, G. W. and R. A. Stockey. 1989. Fossil Ophioglos-
sales in the Paleocene of western North America. Am.
J. Bot. 76: 637-644.
. 1991. Onoclea sensibilis in the Paleocene of North
America, a dramatic example of structural and ecolog-
ical stasis. Rev. Palaeobot. Palynol. 70:113-124.
Rothwell, G. W. and R. A. Stockey. 1994. The role of Hy-
dropteris pinnata Gen. et Sp. Nov. in reconstructing
the cladistics of heterosporous ferns. Am. J. Bot. 81:
479-492.
Rouse, G. E. and S. K. Srivastava. 1972. Palynological zona-
tion of Cretaceous and Early Tertiary rocks of the Bon-
net Plume Formation, northeastern Yukon, Canada.
Can. J. Earth Sci. 9:1163-1179.
Savin, S. M. 1977. The history of the Earth's surface tem-
perature during the past 100 million years. Annu. Rev.
Earth Planet. Sci. 5: 319-355.
Scotese, C. R., L. M. Gahagan, and R. L. Larson. 1989. Plate
tectonic reconstructions of the Cretaceous and Ceno-
zoic ocean basins. In: C. R. Scotese and W. W. Sager
(eds.), Mesozoic and Cenozoic Plate Reconstructions.
Elsevier Science Publishers, Amsterdam, pp. 27-48.
Scotese, C. R. and W. W. Sager (eds.). 1989. Mesozoic and
Cenozoic Plate Reconstructions. Elsevier Science Pub-
lishers, Amsterdam.
Sellwood, B. W., G. D. Price, and P. J. Valdes. 1994. Cooler
estimates of Cretaceous temperatures. Nature 370:
453-455.
Serbet, R. and R. A. Stockey. 1991. Taxodiaceous pollen
cones from the Upper Cretaceous (Horseshoe Canyon
Formation) of Dmmheller, Alberta, Canada. Rev.
Palaeobot. Palynol. 70: 67-76.
Sheehan, R M. and D. E. Fastovsky. 1992. Major extinctions
of land-dwelling vertebrates at the Cretaceous -
Tertiary boundary, eastern Montana. Geology 20:
556-560.
Sloan, L. C. and E. J. Barron. 1990. "Equable" climates dur-
ing Earth history? Geology 18: 489-492.
Sloan, L. C. and D. K. Rea. 1995. Atmospheric carbon diox-
ide and early Eocene climate: a general circulation
modeling sensitivity study. Palaeogeogr., Palaeoclima-
tol., Palaeoecol. 119: 275-292.
Smiley, C. J. 1967. Paleoclimatic interpretations of some
Mesozoic floral sequences. Am. Assoc. Petrol. Geol.
Bull. 51: 849-863.
Snead, R. G. 1969. Microfloral diagnosis of the Creta-
ceous-Tertiary boundary, central Alberta. Res. Council
Alberta Bull. 25: 1-148.
Spicer, R. A., R. J. Burnham, R Grant, and H. Glicken. 1985.
Pityrogramma calomelanos, the primary, post-
eruption colonizer of Volcan Chichonal, Chiapas, Mex-
ico. Am. Fern J. 75: 1-5.
Spicer, R. A., K. S. Davies, and A. B. Herman. 1994. Cir-
cum-Arctic plant fossils and the Cretaceous-Tertiary
transition. In: M. C. Boulter and H. C. Fisher (eds.),
Cenozoic Plants and Climates of the Arctic. NATO ASI
Series 1: Global Environmental Change. Springer -
Verlag, Berlin. Vol. 27, pp. 161-174.
Spicer, R. A. and J. T. Parrish. 1986a. Paleobotanical evi-
dence for cool north polar climates in Middle Creta-
ceous (Albian-Cenomanian) time. Geology 14: 703-
706.
Late Cretaceous through Early Eocene History 183
—. 1986b. Fossil woods from northern Alaska and cli-
mate near the Middle Cretaceous North Pole. Geol.
Soc. Am. Abstr. Program 18: 759.
-. 1990. Late Cretaceous-Early Tertiary palaeocli-
mates of northern high latitudes: a quantitative view. J.
Geol. Soc. Lond. 147: 329-341.
Srivastava, S. K. 1970. Pollen biostratigraphy and paleoe-
cology of the Edmonton Formation (Maestrichtian),
Alberta, Canada. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 7: 221-276.
. 1981. Evolution of Upper Cretaceous phytogeo-
provinces and their pollen flora. Rev. Palaeobot. Paly-
nol. 35: 155-173.
. 1994a. Evolution of Cretaceous phytogeoprovinces,
continents and climates. Rev. Palaeobot. Palynol. 82:
197-224.
-. 1994b. Palynology of the Cretaceous-Tertiary
boundary in the Scollard Formation of Alberta,
Canada, and global KTB events. Rev. Palaeobot. Paly-
nol. 83: 137-158.
Standley, E. A. 1965. Upper Cretaceous and Paleocene
plant microfossils and Paleocene dinoflagellates and
hystrichosphaerids from northwestern South Dakota.
Bull. Am. Paleontol. 49: 1-384.
Stebbins, G. L., Jr. 1974. Flowering Plants: Evolution above
the Species Level. Belknap Press, Harvard University.
Cambridge, MA.
Stockey, R. A., G. L. Hoffman, and G. W. Rothwell. 1997.
The fossil monocot Limnobiophyllum scutatum: re-
solving the phylogeny of Lemnaceae. Am. J. Bot. 84:
355-368.
Stockey, R. A. and S. R. Manchester. 1988. A fossil flower
with in situ Pistillipollenites from the Eocene of
British Columbia. Can. J. Bot. 66: 313-318.
Stockey, R. A. and G. W. Rothwell. 1997. The aquatic an-
giosperm Trapago angulata from the Upper Creta-
ceous (Maastrichtian) St. Mary River Formation of
southern Alberta. Internat. J. Plant Sci. 158: 83-94.
Stucky, R. K. and L. Krishtalka. 1990. Revision of the Wind
River faunas, Early Eocene of central Wyoming. Part
10. Bunophorus (Mammalia, Artiodactyla). Ann.
Carnegie Mus. 59: 149-171.
and A. D. Redline. 1990. Geology, vertebrate fauna,
and paleoecology of the Busk Spring quarries (Early
Eocene, Wind River Formation), Wyoming. In: T. M.
Bown and K. D. Rose (eds.), Dawn of the Age of Mam-
mals in the Northern Part of the Rocky Mountain Inte-
rior, North America. Geological Society of America,
Boulder, CO. Special Paper 243, pp. 169-186.
Sun, E and R. A. Stockey. 1992. A new species of Palaeo-
carpinus (Betulaceae) based on infructescences, fruits,
and associated staminate inflorescences and leaves
from the Paleocene of Alberta, Canada. Int. J. Plant Sci.
153: 136-146.
Sweet, A. R. 1978. Palynology of the Ravenscrag and
Frenchman formations. In: J. A. Irvin, S. H. Whitaker,
and P. L. Broughton (eds.), Coal Resources of Southern
Saskatchewan: A Model for Evaluation Methodology.
Geological Survey of Canada, Ottawa, Economic Geo-
logical Report. Vol. 30, pp. 29-38.
, D. R. Braman, and J. E Lerbekmo. 1990. Palynofloral
response to K/T boundary events; a transitory inter-
ruption within a dynamic system. In: V. L. Sharpton
and R D. Ward (eds.), Global Catastrophes in Earth His-
184 Late Cretaceous and Cenozoic History of North American Vegetation
tory. Geological Society of America, Boulder, CO. Spe-
cial Paper 247, pp. 457-469.
Taylor, D. W. and W. L. Crepet. 1987. Fossil floral evidence
of Malpighiaceae and an early plant-pollinator rela-
tionship. Am. J. Bot. 74: 274-286.
Thelin, G. R and R. J. Pike. 1991. Landforms of the Conter-
minous United States—A Digital Shaded-Relief Por-
trayal. U.S. Geological Survey Miscellaneous Investi-
gation Series Map 1-2206. U.S. Geological Survey,
Reston, VA.
Thompson, S. L. and E. J. Barren. 1981. Comparison of Cre-
taceous and present Earth albedos: implications for
the causes of paleoclimates. J. Geol. 89: 143-167.
Tiffney, B. H. 1984. Seed size, dispersal syndromes, and the
rise of the angiosperms: evidence and hypothesis.
Ann. Missouri Bot. Garden 71: 551-576.
. 1985. The Eocene North Atlantic land bridge: its
importance in Tertiary and modern phytogeography of
the northern hemisphere. J. Arnold Arb. 66: 243-273.
-. 1994. An estimate of the early Tertiary paleocli-
mate of the southern Arctic. In: M. C. Boulter and H. C.
Fisher (eds.), Cenozoic Plants and Climates of the Arc-
tic. NATO ASI Series 1: Global Environmental Change.
Springer-Verlag, Berlin. Vol. 27, pp. 267-295.
Traverse, A. 1988. Paleopalynology. Unwin Hyman,
Boston.
Triplehorn, D. M., D. L. Turner, and C. W. Naeser. 1984. Ra-
diometric age of the Chickaloon Formation of south-
central Alaska: location of the Paleocene-Eocene
boundary. Geol. Soc. Am. Bull. 95: 740-742.
Tschudy, R. H. 1970 (1971). Palynology of the Cretaceous-
Tertiary boundary in the northern Rocky Mountain
and Mississippi Embayment regions. In: R. M.
Kosanke and A. T. Cross (eds.), Symposium on Pal-
ynology of the Late Cretaceous and Early Tertiary. Ge-
ological Society of America, Boulder, CO. Special
Paper 127, pp. 65-112.
. 1975. Normapolles pollen from the Mississippi
Embayment. U.S. Geological Survey, Reston, VA. Pro-
fessional Paper 865, pp. 1-42.
. 1980. Normapolles pollen from Aquipollenites
province, western United States. N. Mex. Bur. Mines
Min. Resources Circ. 170: 1-14.
. 1981. Geographic distribution and dispersal of
Normapolles genera in North America. Rev. Palaeobot.
Palynol. 35: 283-314.
-, C. L. Pillmore, C. J. Orth, J. S. Gilmore, and J. D.
Knight. 1984. Disruption of the terrestrial plant
ecosystem at the Cretaceous-Tertiary boundary, West-
ern Interior. Science 225: 1030-1032.
Tselioudis, G., A. A. Lacis, D. Rind, and W. B. Rossow.
1993. Potential effects of cloud optical thickness on
climate warming. Nature 366: 670-672.
Upchurch, G. R., Jr. 1989. Terrestrial environmental
changes and extinction patterns at the Cretaceous-Ter-
tiary boundary, North America. In: S. K. Donovan
(ed.), Mass Extinctions: Processes and Evidence. Co-
lumbia University Press, New York. pp. 195-216.
. 1995. Dispersed angiosperm cuticles: their history,
preparation, and application to the rise of angiosperms
in Cretaceous and Paleocene coals, southern Western
Interior of North America. Int. J. Coal Geol. 28:
161-227.
and D. L. Dilcher. 1990. Cenomanian angiosperm
leaf megafossils, Dakota Formation, Rose Creek local-
ity, Jefferson County, southeastern Nebraska. U.S. Geol.
Surv. Bull. 1915: 1-55.
Upchurch, G. R., Jr. and J. A. Wolfe. 1987. Mid-Cretaceous
to Early Tertiary vegetation and climate: evidence from
fossil leaves and woods. In: E. M. Friis, W. G. Chaloner,
and R R. Crane (eds.), The Origins of Angiosperms and
Their Biological Consequences. Cambridge University
Press, Cambridge, U.K. pp. 75-105.
. 1993. Cretaceous vegetation of the Western Interior
and adjacent regions of North America. In: W. G. E.
Caldwell and E. G. Kauffman (eds.), Evolution of the
Western Interior Basin. Geological Association of Can.,
St. Johns, Newfoundland, Special Paper 39, pp. 243-
281.
Van Valen, L. M. and R. E. Sloan. 1977. Ecology and extinc-
tion of the dinosaurs. Evol. Theory 2: 37-64.
Vogt, P. R. 1979. Global magmatic episodes: new evidence
and implications for the steady-state mid-oceanic
ridge. Geology 7: 93-98.
Webb, S. D. 1989. The fourth dimension in North American
terrestrial mammal communities. In: D. W. Morris, Z.
Abramsky, B. J. Fox, and M. R. Willig (eds.), Patterns in
the Structure of Mammalian Communities. Museum of
Texas Tech University, Lubbock, TX. Special Paper 28,
pp. 181-203.
West, R. M. and M. R. Dawson. 1978. Vertebrate paleontol-
ogy and the Cenozoic history of the North Atlantic re-
gion. Polarforschung 48:103-119.
Wheeler, E. A. 1991. Paleocene dicotyledonous trees from
Big Bend National Park, Texas: variability in wood
types common in the Late Cretaceous and Early Ter-
tiary, and ecological inferences. Am. J. Bot. 78:
658-671.
and P. Baas. 1991. A survey of the fossil record for
dicotyledonous wood and its significance for evolu-
tionary and ecological wood anatomy. Int. Assoc.
Wood Anat. Bull. 12: 275-332.
Wheeler, E. A., M. Lee, and L. C. Matten. 1987. Dicotyledo-
nous woods from the Upper Cretaceous of southern
Illinois. Bot. J. Linn. Soc. 95: 77-100.
Wheeler, E. A., T. M. Lehman, and P. E. Gasson. 1994.
Javelinoxylon, an Upper Cretaceous dicotyledonous
tree from Big Bend National Park, Texas, with pre-
sumed malvalean affinities. Am. J. Bot. 81: 703-710.
Wheeler, E. A., R. A. Scott, and E. S. Barghoorn. 1977. Fos-
sil dicotyledonous woods from Yellowstone National
Park. J. Arnold Arb. 58: 280-302.
. 1978. Fossil dicotyledonous woods from Yellow-
stone National Park, II. J. Arnold Arb. 59: 1-26.
Wing, S. L. 1980. Fossil floras and plant-bearing beds of the
central Bighorn Basin. In: R D. Gingerich (ed.), Early
Cenozoic Paleontology and Stratigraphy of the
Bighorn Basin. University of Michigan Museum Pale-
ontol., Ann Arbor, MI, 119-125.
. 1981. A Study of Paleoecology and Paleobotany in
the Willwood Formation (Early Eocene, Wyoming).
Ph.D. dissertation, Yale University, New Haven, CT.
. 1984. Relation of paleovegetation to geometry and
cyclicity of some fluvial carbonaceous deposits. J. Sed-
iment. Petrol. 54: 52-66.
. 1987. Eocene and Oligocene floras and vegetation
of the Rocky Mountains. Ann. Missouri Bot. Garden
74: 748-784.

. 1996. Paleocene-Eocene floral and climate change
in the northern Rocky Mountains, USA. International
Organization of Palaeobotany Fifth Quadrennial Con-
ference, Santa Barbara, CA, June-July, 1996. p. 113.
Wing, S. L. 1997. Global warming and plant species rich-
ness: a case study of the Paleocene/Eocene boundary.
In: M. L. Reaka-Kudla, D. E. Wilson, and E. O. Wilson
(eds.), Biodiversity II. Understanding and Protecting
our Natural Resources. Joseph Henry Press, Washing-
ton, D.C. pp. 163-185.
Wing, S. L., J. Alroy, and L. J. Hickey. 1995. Plant and mam-
mal diversity in the Paleocene to early Eocene of the
Bighorn Basin. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 115: 117-155.
and T. M. Bown. 1985. Fine scale reconstruction of
late Paleocene-Early Eocene paleogeography in the
Bighorn Basin of northern Wyoming. In: R. M. Flores
and S. S. Kaplan (eds.), Cenozoic Paleogeography of
the West-Central United States. Rocky Mountain Sec-
tion, Society of Economic Paleontology Mineral.,
Golden, CO. pp. 93-106.
and J. D. Obradovich. 1991. Early Eocene biotic and
climatic change in interior western North America.
Geology 19: 1189-1192.
Wing, S. L. and D. R. Greenwood. 1993. Fossils and fossil
climate: the case for equable continental interiors.
Phil. Trans. R. Soc. Lond. B 341: 243-252.
Wing, S. L. and L. J. Hickey. 1984. The Platycarya perplex
and the evolution of the Juglandaceae. Am. J. Bot. 71:
388-411.
Wing, S. L. and B. H. Tiffney. 1987. Interactions of an-
giosperms and herbivorous terapods through time. In:
E. M. Friis, W. G. Chaloner, and E R. Crane (eds.), The
Origins of the Angiosperms and Their Biological Con-
sequences. Cambridge University Press, Cambridge,
U.K. pp. 203-224.
Wolfe, J. A. 1966. Tertiary plants from the Cook Inlet Re-
gion, Alaska. U.S. Geological Survey, Reston, VA. Pro-
fessional Paper 398-B, pp. 1-32.
. 1972. An interpretation of Alaskan Tertiary floras.
In: A. Graham (ed.), Floristics and Paleofloristics of
Asia and Eastern North America. Elsevier Science
Publishers, Amsterdam, pp. 201-233.
. 1977. Paleogene floras from the Gulf of Alaska Re-
gion. U.S. Geological Survey, Reston, VA. Professional
Paper 997, pp. 1-108.
. 1978. A paleobotanical interpretation of Tertiary
climates in the Northern Hemisphere. Am. Sci. 66:
694-703.
. 1979. Temperature Parameters of Humid to Mesic
Forests of Eastern Asia and Relation to Forests of Other
Regions of the Northern Hemisphere and Australasia.
U.S. Geological Survey, Reston, VA. Professional Paper
1106, pp. 1-37.
. 1980. Tertiary climates and floristic relationships at
high latitudes in the northern hemisphere. Palaeo-
geogr., Palaeoclimatol., Palaeoecol. 30: 313-323.
. 1981a. Paleoclimatic significance of the Oligocene
and Neogene floras of the northwestern United States.
In: K. J. Niklas (ed.), Paleobotany, Paleoecology, and
Evolution. Praeger, New York. Vol. 2, pp. 79-101.
. 1981b. A chronologic framework for Cenozoic
megafossil floras of northwestern North America and
its relation to marine geochronology. Geological Soci-
Late Cretaceous through Early Eocene History 185
ety of America, Boulder, CO. Special Paper 184, pp.
39-47.
—. 1985. Distribution of major vegetational types dur-
ing the Tertiary. In: E. T. Sundquist and W. S. Broecker
(eds.), The Carbon Cycle and Atmospheric CO2: Nat-
ural Variations Archean to Present. Geophysical
Monograph 32. American Geophysical Union, Wash-
ington, D.C. pp. 357-375.
—. 1987. Late Cretaceous-Cenozoic history of decid-
uousness and the terminal Cretaceous event. Paleobi-
ology 13: 215-226.
—. 1990. Palaeobotanical evidence for a marked tem-
perature increase following the Cretaceous/Tertiary
boundary. Nature 343:153-156.
—. 1991. Palaeobotanical evidence for a June "impact
winter" at the Cretaceous/Tertiary boundary. Nature
352:420-423.
—. 1994. Alaskan Palaeogene climates as inferred from
the CLAMP database. In: M. C. Boulter and H. C.
Fisher (eds.), Cenozoic Plants and Climates of the Arc-
tic. NATO ASI Series 1: Global Environmental Change.
Springer-Verlag, Berlin. Vol. 27, pp. 223-237.
—. In press. Relations of environmental change to an-
giosperm evolution during the Late Cretaceous and
Tertiary. In: I. Motomi and H. Nishida (eds.), Prospect
for Systematic Botany in the 21st Century. Springer-
Verlag, Berlin.
and S. McCoy. 1984. Comments and Replies on
"Model for the origin of the Yakutat block, an accreting
terrane in the northern Gulf of Alaska." Geology 12:
564-565.
Wolfe, J. A. and T. Tanai. 1987. Systematics, phylogeny, and
distribution of Acer (maples) in the Cenozoic of west-
ern North America. J. Fac. Sci. Hokkaido Univ. 22:
1-246.
Wolfe, J. A. and G. R. Upchurch, Jr. 1986. Vegetation, cli-
matic and floral changes at the Cretaceous-Tertiary
boundary. Nature 324: 148-152.
. 1987a. North American nonmarine climates and
vegetation during the Late Cretaceous. Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 61: 33-77.
-. 1987b. Leaf assemblages across the Cretaceous-
Tertiary boundary in the Raton Basin, New Mexico
and Colorado. Proc. Natl. Acad. Sci. USA 84:
5096-5100.
Ziegler, A. M., M. L. Hulver, A. L. Lottes, and W. E
Schmachtenberg. 1984. Uniformitarianism and
palaeoclimates: inferences from the distribution of car-
bonate rocks. In: P. Brenchley (ed.), Fossils and Cli-
mate. Wiley, New York. pp. 3-25.
General Readings: K-T Boundary
Archibald, J. D. 1981. The earliest known Palaeocene mam-
mal fauna and its implications for the Cretaceous -
Tertiary transition. Nature 291: 650-652. (An example
of an earlier essay arguing for gradual, and presumably
climatically controlled, transition and extinctions at
the K-T boundary.)
Lerbekmo, J. F, A. R. Sweet, and R. M. St. Louis. 1987. The
relationship between the iridium anomaly and paly-
nological floral events at three Cretaceous-Tertiary
186 Late Cretaceous and Cenozoic History of North American Vegetation

boundary localities in western Canada. Geol. Soc. Am. . 1997. Upstart ice age theory gets attentative but
Bull. 99: 325-330. chilly hearing. Science 277: 183-184.
Mclver, E. E., A. R. Sweet, and J. E Basinger. 1991. Sixty- McGowran, B. 1990. Fifty million years ago. Sci. Am. 78:
five-million-year-old flowers bearing pollen of the ex- 30-39.
tinct triprojectate complex—a Cretaceous-Tertiary Morell, V. 1993. How lethal was the K-T impact? Science
boundary survivor. Rev. Palaeobot. Palynol. 70: 77-88. 261: 1518-1519.
Pillmore, C. L., R. H. Tschudy, C. J. Orth, J. S. Gilmore, and . 1994. Warm-blooded dino debate blows hot and
J. D. Knight. 1984. Geologic framework of nonmarine cold. Science 265: 188.
Cretaceous-Tertiary boundary sites, Raton Basin, New Roush, W. 1996. Probing flower's genetic past. Science 273:
Mexico and Colorado. Science 223: 1180-1183. (See 1339-1340.
also three adjacent essays in this issue on the same
subject.)
Sweet, A. R. 1986. The Cretaceous-Tertiary boundary in Notes
the central Alberta foothills: II. Miospore and pollen 1. On a historical note, samples from the McNairy Sand
taxonomy. Can. J. Earth Sci. 23: 1375-1388. [For part I (Ripley) Formation in Tennessee were part of the original
see Jerzykiewicz and Sweet (1986).] material used in the development of the oxygen isotope
technique for paleotemperature analysis (Lowenstam and
General Readings Epstein, 1954). Their estimate of MAT, made over 40 years
ago, was 20-29°C.
Barren, E. J. 1994. Chill over the Cretaceous. Nature 370: 2. The Hell Creek Formation consists of sediments de-
415. posited ~2 Ma before the terminal Cretaceous extinctions
Chaloner, W. G. and A. Hallam (eds.). 1989. Evolution and at 65 Ma. The overlying Tullock Formation was deposited
Extinction. Cambridge University Press, Cambridge, immediately afterward. For an interesting discussion of the
U.K. paleontological analyses at this important site regarding
Dorf, E. 1964. The petrified forests of Yellowstone Park. the bolide impact, see Sheehan and Fastovsky (1992) and
Sci. Am. 210:106-114. the reference to Kerr (1992a).
Kerr, R. A. 1991. Did a burst of volcanism overheat ancient 3. The paleontological literature on K-T boundary
Earth? Science 251: 746-747. events is extensive and cannot be reviewed in detail here.
. 1992a. Did an asteroid leave its mark in Montana For those interested in reading further on this subject, a
bones? Science 256: 1395-1396. supplementary list of essays with emphasis on paleo-
. 1992b. When climate twitches, evolution takes botany and paleopalynology is provided under General
great leaps. Science 257:1622-1624. Readings: K-T Boundary in the References section.
. 1993a. Fossils tell of mild winters in an ancient 4. Oswald Heer (1883) described the extensive fossil
hothouse. Science 261: 82. floras of the Arctic in a seven-volume work published be-
. 1993b. A bigger death knell for the dinosaurs? Sci- tween 1868 and 1883 [For an account of his life see An-
ence 261: 1518-1519. drews (1980).]
 SIX
Middle Eocene through Early Miocene
North American Vegetational History
50-16.3 Ma
CONTEXT SUMMARY
During the Middle Eocene through the Early Miocene, ero-
sion of the Appalachian Mountains exceeded uplift and
there was a net reduction in elevation (Figs. 5.19, 6.1). In
the Rocky Mountains uplift continued through the Middle
Eocene (end of the Laramide orogeny), waned in the Mid-
dle Tertiary, and then increased beginning at about 10 Ma.
Earlier reconstructions placed paleoelevations in the
Rocky Mountains during the Middle Eocene through the
Early Miocene at approximately half the present relief. The
maximum elevation in the Front Ranges during the latest
Eocene was estimated at -2500 m (~8000 ft; MacGinitie,
1953). Recent approximations are for nearly modern eleva-
tions in several areas by the Eocene-Oligocene. Extensive
Eocene volcanism deposited ash and blocked drainage sys-
tems, augmenting uplift and facilitating the preservation of
extensive fossil floras and faunas. In the far west the begin-
ning of Tertiary volcanism in the Sierra Nevada is dated at
-33 Ma near the Eocene-Oligocene boundary.
A drying trend becomes evident in the Middle Eocene
and reduced moisture, along with the waning of volcanic
activity in the Oligocene, restricted conditions favorable to
fossilization. The number of Oligocene floras in the north-
ern Rocky Mountains is considerably fewer than in younger
deposits to the west.
In the absence of extensive plate reorganization and
orogeny, CO2 concentration decreased, which contributed
to a temperature decline that continued through the Ceno-
zoic and intensified in the Late Tertiary. Recall from Chap-
ter 2 (sections on orogeny and volcanism) that uplift plays
a role in determining long-term climate by creating rain-
shadows, altering atmospheric circulation patterns, and
increasing the erosion of silicate rocks that causes a draw-
down of CO2. This allows heat to escape from the tropo-
sphere and results in lower temperatures. Marine benthic
187
temperatures were ~10°C in the early Late Eocene and ~2°C
near the Eocene-Oligocene boundary, assuming an essen-
tially ice-free Earth during that time, and increased to
~5-6°Cnear the end of the Early Miocene (Fig. 3.1). Tem-
peratures over land in the midnorthern latitudes are esti-
mated to have dropped by ~12°C between the Late Eocene
and Early Oligocene (Wolfe, 1992a). The annual range in
temperature also widened from an estimated 3-5°C in the
Pacific Northwest in the Middle Eocene to 25°C in the
Oligocene. From the Middle Eocene through the Late
Eocene the decline correlates with decreasing CO2 concen-
tration; after that time it was additionally influenced in the
high latitudes by full opening of the North Atlantic seaway,
which allowed an intermingling of Arctic and Atlantic
Ocean waters with the formation of NADW.
In the southern hemisphere there was also a change in
ocean circulation in the Oligocene caused by partial open-
ing of the Drake Passage (55-50 Ma) and the further sepa-
ration of Australia and Tasmania from Antarctica (30-25
Ma). This led to the development of a circum-Antarctic
current that thermally isolated Antarctica from warmer
equatorial waters and contributed to continental glacia-
tion. Glacial deposits as old as the Middle Eocene have
been reported from King George Island off the Antarctic
Peninsula (Birkenmajer, 1987). The oldest unequivocal ev-
idence for a substantial continental ice sheet on Antarctica
is the Early Oligocene (chron C13N; 33.5 Ma) in the form
of ice-rafted detritus on the East Antarctic margin of the
Weddell Sea (ODP Leg 113) and on the southern and cen-
tral parts of the Kerguelen Plateau (ODP Leg 119, 120).
This correlates with a 1% increase in 18O (Berggren and
Prothero, 1992). Among the results were an enhancement
of cold deep waters, steeper pole to equator temperature
gradients, and strengthened high-pressure systems with
concomitant increased and seasonal dryness. During the
Oligocene to the end of the Early Miocene temperatures
188 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 6.1. Principal fossil floras mentioned
in the text plotted with general reference to the paleo-
temperature record (lower curve) and sea-level record
(upper curve) for the Middle Eocene through
the Early Miocene.
fluctuated near the line between glacial and ice-free condi-
tions (Fig. 3.1; Zachos et al., 1997). The Miocene thermal
optimum occurred between -21 and 15 Ma, corresponding
to planktic foraminifera zones N5-N10. In the Middle
Miocene (15-4 Ma) another temperature drop ushered in
extensive continental glaciations on Antarctica and early
lowland glaciation in the Arctic.
Sea level fell during the interval between the Middle
Eocene and Early Miocene, and the change shows a pattern
consistent with temperatures fluctuating along the bound-
ary between intermittent glacial and ice-free conditions. A
core from near Mays Landing on the coastal plain of New
Jersey preserves erosion surfaces near the Lower Eocene-
Middle Eocene boundary, near the Middle Eocene-Upper
Eocene boundary, within the Middle Eocene, and in the
Middle Oligocene (Owens et al., 1988). The most signifi-
cant sea-level event was in the Oligocene at 31-29 Ma.
Other lowerings occurred at 21, 16.5, and 15.5 Ma (Fig.
3.6). These changes correlate with hiatuses in other conti-
nental shelf, slope, and epicontinental settings and are
therefore suggestive of eustatic changes in sea level (Miller
et al., 1990). For example, the fall of about 30 m at 31-29
Ma correlates approximately with the evidence from ice-
rafted detritus of increasing ice on Antarctica.
The epicontinental sea continued to retreat from mid-
continent North America and by the end of the Early

Miocene had reached a point about halfway between the
Middle Eocene position shown in Fig. 5.6 and the present
Gulf Coast (viz., along a line running midway between
Austin and Houston, Texas, to just south of Jackson, Mis-
sissippi, and around the southern terminus of the Appa-
lachian Mountains). The depositional setting of a gradually
retreating shoreline bordered by swamps and lakes facili-
tated the preservation of extensive fossil floras of coastal
and lowland vegetation, and pollen was also being blown
in from the adjacent Appalachian highlands. The retreat
resulted in an increased albedo, an increase in the erosion
of silicate and other rocks, and greater continentality in the
interior regions.
Precipitation decreased and became more seasonal. Pale-
osols (fossil soil horizons) from the Badlands of South
Dakota show a drying trend from an estimated annual pre-
cipitation of 1000 mm in the early Late Eocene (38 Ma), to
500-900 mm in the Early Oligocene (32 Ma), to 450-500
mm at 30.5 Ma, and to 250-450 mm in the Late Oligocene-
Early Miocene (29.5 Ma; Retallack, 1983a,b, 1990).
Some Ir anomalies have been reported from the Late
Eocene (-35 to -40 Ma; Hut et al., 1987) and from the Late
Miocene (-10 Ma; Asaro et al., 1988); but the concentra-
tions are much lower than at the K-T boundary and they
do not correspond to extinction events in the marine
realm, which are presently viewed as mostly endogenic
rather than exogenic in origin. Microtektites are associated
only with a minor extinction event at Chron C15R (-36.5
Ma). Impact structures of Late Eocene age (35.5 Ma) have
been discovered in the Chesapeake Bay region and in
Siberia (Chapter 2), but the biological consequences of the
event have not been assessed (Koeberl et al., 1996).
Summarizing events during the Paleogene is compli-
cated by the fact that the termination date for the Eocene
has been the most unsettled of all the Tertiary epochs. Esti-
mates for the Eocene-Oligocene boundary range from 38
Ma to as late as 32 Ma (Wolfe, 1981a), but results from a re-
cently developed single crystal laser fusion (SCLF) 40Ar/39Ar
radiometric dating technique are focusing on an age of -34
Ma (Berggren et al., 1995; 33.5 Ma, Prothero, 1994):
The Eocene/Oligocene boundary was recently stratotyp-
ified at the 19m level in the Massignano section of the
northern Apennines. It is denoted by the last occurrence
(LO) of the planktonic foraminiferal genus Hantkenina
which occurs at a stratigraphic level in Magnetozone
C13R, estimated to lie at a level 0.14 of the interpolated
distance between the base of C13N and the top of C15N,
with an estimated numerical age of 34.0 Ma. (Prothero
and Berggren, 1992, p. 3)
New dates for other epoch-stage boundaries and fossil
biotas are also being published, and erroneous dates are
being corrected. The movement of epoch boundaries can
affect the age name used for assemblages. For example, if a
fossil flora is correctly dated at 35 Ma, and the Eocene-
Oligocene boundary is placed at 36 Ma, the flora belongs to
Middle Eocene through Early Miocene History 189
the Oligocene. If the boundary is placed at 34 Ma, the flora
then belongs to the Eocene. Similar situations arise if the
boundary is established but new and different dates are ob-
tained for the flora. Both kinds of age "changes" are partly
semantic, and a flora can be found under different epoch
names in the literature. In the following chapters new
dates (some published and others unpublished, in press, or
estimated and/or extrapolated) are indicated in brackets
along with previously published dates.
In discussing the climates of the Early Eocene, it was
noted that the trend toward maximum warmth was accom-
plished in abrupt, steep, short-term increments. Accumu-
lating evidence suggests that this mode is characteristic of
long-term trends generally, and that climate jumps from
one stable state or threshold to another in short pulses. The
temperature decline from the Middle Eocene through the
Early Miocene fell in the same stepwise fashion. This pe-
riod is one of the most complex in North American vegeta-
tional history and also one of the more challenging to out-
line because of previous uncertainty as to when the decline
began and a resulting confused timetable.
The phrase "Oligocene deterioration" was initially used
to characterize a temperature decline and associated vege-
tation changes evident in fossil floras from Alaska, the Pa-
cific Northwest, and the Gulf Coast (Wolfe, 1971). Subse-
quently, these changes were labeled the "Terminal Eocene
Event" (Wolfe, 1978), implying that there was a single
event and that it occurred at the end of the Eocene. The
term became entrenched in the literature, in part, because
of the title of a book, Terminal Eocene Events (Pomerol and
Premoli-Silva, 1986), resulting from a UNESCO sponsored
project, "Geological Events at the Eocene-Oligocene
Boundary" as part of the International Geological Correla-
tion Program. Ironically, among the conclusions in the
book were "1) the Eocene-Oligocene boundary considered
as a major evolutionary break of the Cenozoic is character-
ized by a gradual, and not abrupt, environmental change
which started in the Middle Eocene and continued into the
earliest Oligocene for some 300,000 years"; that is, it began
in the Middle Eocene rather than at the end of the Eocene
and extended over a span of about 10 Ma, and "2) There is
not a single terminal Eocene event, but a series of events
concentrated close to the Eocene-Oligocene boundary as
testified by an acceleration of the rate of overturn among
most of the taxonomic groups, by an intensification of the
climatic deterioration and by a major reorganization of the
oceanic water masses" (Pomerol and Premoli-Silva, 1986,
p. vi; see also Pomerol, 1981).
Evidence presented in 1989 at a Penrose Conference
and a Theme Session of the Geological Society of America
(Prothero and Berggren, 1992) confirm that the most signif-
icant cooling event in the Paleogene began in the Middle
Eocene (Bartonian-Priabonian boundary; see also Axel-
rod, 1992; Marty et al., 1988), was augmented by a slighter
cooling at the Eocene-Oligocene boundary and by more
pronounced cooling in the mid-Oligocene (32.5 Ma in the
190 Late Cretaceous and Cenozoic History of North American Vegetation
revised chronology used by Berggren and Prothero, 1992;
Fig. 3.1), and occurred in a stepwise fashion. Eocene floras
from southern England show a similar pattern (Collinson
et al., 1981). Planktonic foraminifera from deep-sea sites
record five extinction events between the Lutetian and the
Late Rupelian, as summarized by Prothero (1989; new
dates fide Prothero, personal communication, 1996):
The Lutetian-Bartonian (Middle Eocene) event oc-
curred at the boundary between planktonic foraminifera
zones PI2 and Pi3, at the top of magnetic Chron C18R
(-41 Ma). This was a minor event near the beginning of
the cooling trend of the Middle Eocene and corresponds to
a drop in sea level termed Tejas A3.4 by Haq et al. (1987;
Fig. 6.1).
The Bartonian-Priabonian (Middle-Late Eocene) event
occurred at the boundary at the P14-P15 boundary at
Chron C17R (~37 Ma). This was a major event involving
the extinction of spinose tropical foraminifera and migra-
tion toward the equator of midlatitude species. Among the
calcareous nanoplankton,1 nearly 50% of the species dis-
appeared. A cooling event is also evident in the oxygen iso-
tope record, and a major drop in sea level occurred at the
top of Tejas A3.6 (Fig. 6.1).
The Late Priabonian event (Late Eocene) occurred at the
P15-P16 boundary at Chron C15R (-35 Ma). This was a
minor event that involved continued cooling, lowering of
sea level, and the only extraterrestrial impacts.
The Terminal Eocene Event is now mostly considered
as Early Oligocene, P17-P18 boundary at mid-Chron
C13R (-33.5 Ma). Few extinctions are evident among the
planktonic foraminifera and nanoplankton, but more oc-
curred in the ostracodes and benthic foraminifera. Oxygen
isotopes record a cooling of ~2°C and a slight drop in sea
level.
The mid-Oligocene was at the Late Rupelian at the top
of P19 and mid-Chron CllR -30.5 Ma; this signified
Oligocene deterioration (Wolfe, 1971) and was later called
the Terminal Eocene Event (Wolfe, 1978; see also Prothero,
1989). This event essentially eliminated most holdovers
from the warmer Late Eocene and transitional Early
Oligocene as marked by the extinction of many cool-water
foraminifera and nanoplankton, oxygen isotope records,
and a fall in sea level. From -32 through 30 Ma (Tejas A4.5)
the largest sea-level drop in Tertiary history is recorded,
and it likely was the result of the beginning of full conti-
nental glaciation on Antarctica.
Thus, there are five extinction events recorded in the
marine realm (41, 37, 35, 33.5, 30.5 Ma) spread over about
10 m.y, which precludes any single catastrophic event as
the causal mechanism for biotic changes during this pe-
riod. As would be expected, there was a panoply of causes,
the principal ones being continued decline in CO2 concen-
tration as plate reorganization and erosion rates waned and
changes in ocean circulation.
It is not possible to reconstruct the terrestrial plant
record with the same resolution as in the marine realm,
where near-continuous records are frequently available
from DSDP and ODP cores. The plant megafossil record is
particularly discontinuous, but plant microfossils from
wells approximate the sequences from marine cores in
some instances. The terrestrial plant record is summarized
by Frederiksen:
[I]t now appears that, at least in the Gulf Coast, the "ter-
minal Eocene" event took place entirely at the begin-
ning of the Oligocene and that little if any climatic
change can be detected in this region at the end of the
Eocene. (1988, p. 46)
The timing of faunal history for this period is unsettled
because current revisions in the isotopic time scale affect
the NALMAs for the Paleogene (older by -2 m.y.), and
these are currently being recalibrated to the Global Polarity
Time Scale (GPTS; Figs. 3.12, 3.13, 5.19):
The Uintan land mammal "age" (long thought to be
[L]ate Eocene) is now placed in the Middle Eocene, and
the Uintan/Duchesnean boundary appears to correspond
to the [M]iddle/[L]ate Eocene (Bartonian/Priabonian)
boundary [the Uintan and Duchesnean are now both
[M]iddle Eocene fide D. R. Prothero, personal communi-
cation, 1996]. The Chadronian, long thought to be Early
Oligocene, now appears to be [L]ate Eocene, and the
Chadronian/Orellan boundary (at about 34 Ma) appears
to correspond to the Eocene-Oligocene boundary. The
Orellan and Whitneyan, once considered [M]iddle and
Late Oligocene, are now placed in the [E]arly Oligocene.
Since these discoveries are so recent, even faunal sum-
maries like those of Prothero (1985, 1989) and Stiicky
(1990) are now out of date." (Prothero and Berggren,
1992, p. 16; see also Prothero, 1995)
The events between the Middle Eocene and Early
Oligocene culminated in a rapid change in European fau-
nas known as the Grande Coupure (the great break). Sixty
percent of the mammals of Europe became extinct at
about 32.5 Ma (NP22/23; Early Oligocene rather than the
previous assignment to the Eocene-Oligocene bound-
ary). In North America the older protrogomorph rodents
were replaced by modern families (heteromyids, geomy-
ids, castorids, sciurids, cylindrodontids, cricetids). The
first appearances of larger mammals in North America in-
clude Canidae, Mustelidae, Tapiridae, Rhinocerotidae,
Anthracotheriidae, and Tayassuidae spaced through the
Duchesnean and Chadronian. This was due in large part
to invasions across Beringia from the west as latest
Eocene-Early Oligocene glaciers in Antarctica, reaching
up to 70% of their modern volume, drew down sea levels
in a surge that lasted only a few hundred thousand years.
Similarities between Late Eocene and Oligocene Asian
and North American assemblages include, in particular,
groups adapted to open woodland and savanna habitats.
Examples are Mytonolagus (a rabbit); lophodont ro-
dents—such as eomyids, cricetids, and zapodids among
the small herbivores; and piglike entelodonts, crescent-

toothed artiodactyls (Camelidae, Hypertragulidae, Lep-
tomerycidae), and perissodactyls (tapiroids, rhinocero-
tids, chalicotheres) among the larger herbivores (Webb,
1985). The relationship is evident in assemblages from
the Badlands of South Dakota, the Big Bend region of
Texas, and southern California in the small-mammal for-
est faunas of the mid-Eocene to large-mammal woodland
and savanna faunas of the Eocene—Oligocene. The latter
includes herding ungulates (Mesohippus, Leptomeryx,
Poebrotherium, Merycoidodori) and hypsodont small her-
bivores (Palaeolagus, Ischyromys}. The environmental
implication is an expansion of woodland and savanna
habitats resulting from cooling and drying climates. As
might be expected from the Oligocene section of Fig. 3.1,
after the shifts between the Middle Eocene and Early
Oligocene, faunal history during the rest of the Oligocene
was less eventful. The Late Eocene-earliest Oligocene
marks the end of extensive land-mammal interchange
with Europe across the North Atlantic land bridge.
During the Middle Eocene (Early Uintan) a number of
archaic forms disappeared from North America, further re-
ducing prior similarities with European faunas. Lost were
the Condylarthra, Taeniodonta, Tillodontia, Dinocerata,
and northern Notoungulata; and prosimian primates de-
clined (Webb, 1989). About 60% of Early Oligocene genera
were new to North America.
In the Middle Oligocene open country faunas included
Palaeolagus (a rabbit) and the ruminants Leptomeryx, Hy-
pertragulus, and Poebrotherium; stream borders were in-
habited by Protoceras (a relative of the camelids), Elomeryx
(anthracothere), Agnotocastor (a beaver), and Subhyra-
codon (a rhino). Mesohippus (horse) and Merycoidodon
(oreodont) were wide-ranging.
During the Early Miocene (Arikareean NALMA) immi-
grations from Asia, which had been occurring intermit-
tently, increased significantly. They include Cynelos (amph-
icyonid), Menoceras (rhinocerotid), Moropus (chalicothere),
and the smaller Plesiosminthus (zapodid) and Pseudo-
theridomys (eomyid). New introductions from Asia reached
a peak in the Hemingfordian with 16 new genera. These in-
cluded ruminants (pronghorn antilocaprids, giraffe-horned
dromomerycids, hornless moschids, representing forest to
savanna and browsing to grazing mammals); brachypotheri-
ine rhinocerotid ungulates (Teleoceras}; and especially car-
nivores (cats, Pseudaelurus; bears, Ursavus; Hemicyon; and
various species of Cynelos, Cephalogale, Leptarctus, Pota-
motherium, Amphictis, and Mionictis). The principle ro-
dents were beaver (Anchitheriomys), eomyid (Eomys), and
petauristine flying squirrel (Blackia; Webb, 1985). The fau-
nal record documents a fully functioning Bering land bridge
because the animal traffic included such transport modes as
ambulatory, cursorial, amphibious, and volant. The rich and
varied fauna utilizing the bridge is paralleled by equally di-
verse plant assemblages. The faunal evidence suggests that
widespread shrub savanna habitats continued until about 20
Ma, when Miocene grassland savanna began to increase.
Middle Eocene through Early Miocene History 191
MIDDLE THROUGH LATE EOCENE
VEGETATION: 50-35.4 (34) MA
As noted in Chapter 5, the climate of the Late Cretaceous
through the Early Eocene was warm, equable, and mar-
itime. Coastal assemblages show that megathermal condi-
tions (MAT > 20°C) extended to about 48° N (Figs. 6.1, 6.2,
Table 6.1). Areas of microthermal temperature were highly
restricted and at times possibly limited only to mountains
in the polar latitudes (Wolfe, 1987). After the thermal highs
of the Early Eocene, in the Middle Eocene a complex series
of changes began that varied in intensity in different parts
of the continent but trended toward cooler (especially
lower minimum winter temperatures) and more seasonal
winter dry climates. The Middle Eocene represents essen-
tially a transition from the hothouse conditions of the Early
Tertiary to the icehouse conditions of the later Cenozoic.
This is an important time in the modernization of North
American plant communities from ancient Cretaceous and
Early Tertiary predecessors, and several modern plant for-
mations and associations appear during this interval. The
vegetation classification given in Table 1.1 for modern
communities at the formation level can be used for Middle
Eocene and later assemblages.
In discussing vegetational history for the southeastern
United States, provincial stage names are shown in the far-
right column of Fig. 5.20. Middle Eocene through Early
Miocene megafossil floras are known from the Gulf Coast
region, but these have not been revised in recent times; only
certain taxa have been treated through paleomonographic
surveys. In his initial study Berry (1916) listed 341 species
for the "Wilcox" flora, but at the western Kentucky-
Tennessee sites (now assigned to the Middle Eocene Clai-
borne Formation) only a few have been studied recently
(Table 6.2). The prominence of the Leguminosae is partly a
reflection of the climatic changes that were taking place in
the southeastern United States. The megathermal tropical
rain forest of the Paleocene and Early Eocene was being re-
placed by a semideciduous tropical dry forest (Fig. 6.3) in
response to reduced and more seasonally distributed rain-
fall. A corresponding increase in legumes is evident by the
number of specimens having cross-striated pulvini charac-
teristic of the group. The family is presently an important
component of the tropical dry forest of northern Latin
America. Mangroves grew along the coasts, as represented
by pollen of the Old World tropical mangrove palm Nipa
in the Laredo (Claiborne) Formation of south Texas (Gee,
1990; Westgate and Gee, 1990) and Spinizonocolpites
prominanthus on the eastern Gulf Coast thoughout most of
the Eocene (Frederiksen, 1988, pi. 17).
In contrast to the discontinuous and mostly unrevised
megafossil record, pollen and spore assemblages have been
studied extensively in this petroleum-rich area; and much
of the vegetational history is presently based on plant mi-
crofossils (Frederiksen, 1981,1988,1991; Gray, 1960; Jones,
1961; Nichols, 1970). Claiborne palynomorphs include aff.
Table 6.1. Geographic distribution of Middle Eocene through Early Miocene floras of North America.

Eocene
Southeast Claiborne Berry (1916)
Frederiksen (1981, 1988)
(see text)
Jackson Frederiksen (1980a, 1981, 1991)
Mid-Atlantic Mays Landing Owens et al. (1988)
Northeast Lake Hazen Christe (1964), Christe and
(Ellesmere Isl.) Rouse (1976)
West
Wyoming Little Mt, Leopold and MacGinitie (1972)
Tipperary, MacGinitie (1974)
Kisinger Lakes
Colorado-Utah Green River MacGinitie (1969)
Florissant MacGinitie (1953)
Montana: latest Eocene to earliest Oligocene Beaverhead Becker (1969)
Basins
Ruby Becker (1961)
Metzel Ranch Becker (1972)
Morman Creek Becker (1960)
York Ranch Becker (1973)
Idaho Thunder Mt. Axelrod (1990)
Germer Axelrod et al. (1991)
Nevada Elko Axelrod (1992)
Copper Basin Axelrod (1966b)
Bull Run Axelrod (1966b, c)
California Elsinore Frederiksen (1991)
Susanville Wolfe (1978, 1985)
La Porte Potbury (1935),
Wolfe (I992b)
Lower Cedarville
Chalks Bluff MacGinitie (1941)
Oregon Clarno Manchester (1990),
Scott (1954)
Goshen Chaney and Sanborn (1933)
Comstock Sanborn (1935)
Washington Republic Wolfe and Wehr (1987)
British Columbia Princeton Cevallos-Ferriz et al. (1991); Table 6.9 and see text

Oligocene
Southeast
Texas Catahoula Berry (1916); see text
Mississippi Vicksburg Frederiksen (1981)
West
Colorado Creede Axelrod (1987), Wolfe and Schorn (1989, 1990);
Table 6.12
Oregon Bridge Creek Chaney (1924, 1927)
Fossil Manchester and Meyer (1987);
Table 6.13
Rujada Lakhanpal (1958)
Northwest
Mackenzie Delta Richards Norris (1982)
table continued

Figure 6.2, facing page. Landform map of the conterminous United States with overlay of principal Middle to Late
Eocene (1-18), Oligocene (19-23), and Early Miocene (24-27) floras mentioned in the text. (1) Mays Landing; (2)
Wilcox (Claiborne); (3) Laredo (Claiborne); (4) Little Mountain; (5) Kisinger Lakes; (6) (and vicinity) Green River; (7)
Thunder Mountain, Challis Volcanics; (8) Florissant; (9) Copper Basin; (10) Bull Run; (11) Elsinore; (12) Susanville; (13)
La Porte; (14) Comstock, Goshen; (15) Lower Cedarville; (16) Clarno, Bridge Creek, Fossil; (17) Republic; (18) Princeton;
(19) Catahoula; (20) Brandon; (21) Creede; (22) Beaverhead Basins, Metzel Ranch, Mormon Creek, Ruby, York Ranch;
(23) Rujada; (24) Buffalo Canyon; (25) Alvord Creek; (26) Tehachapi; (27) Sutro (Thelin and Pike, 1991).

192
194 Late Cretaceous and Cenozoic History of North American Vegetation

Table 6.1. continued

Early Miocene
Northeast
Vermont Brandon Tiffney (1994), Traverse (1994)
Devon Island Haughton Hickey et al. (1988)
Whitlock and Dawson (1990)
Banks Island Mary Sachs Matthews and Ovenden (1990)
West
Nevada Buffalo Canyon Axelrod (1991)
Sutro Axelrod (1991)
Oregon Alvord Creek Axelrod (1944)
California Tehachapi Axelrod (1939)
Northwest
Alaska Seldovia Point Wolfe and Tanai (1980)

Table 6.2. Megafossils from Middle Eocene Claiborne Acrostichum, Anemia (cf. Mohria], Gramineae, Mpa and
Formation, western Kentucky-Tennessee. other palms, Ephedra (presumably growing in sandy xeric
coastal habitats), Pinus, Alangium, Alnus, probably Ara-
Podocarpaceae
ceae [Proxapertites], Bombacaceae, Betula, Carya, Cas-
Podocarpus (as Taxodium, Dilcher (1969)
Berry, 1916) tanea, Celtis, Eucommia, Fagus, Cyrilla—Cliftonia, Ilex,
Palmae Quercus?, Corylus, Nyssa, Juglans, Platycarya, Pterocarya
Sabal Dilcher (1968), Daghlian and and other Juglandaceae [Momipites], Olacaceae, Sapin-
Dilcher (1972) daceae? [Nuxpollenites], Onagraceae, Ostrya-Carpinus,
Araceae
Platanus, Sapotaceae, Symplocos, Tilia, and Ulmus. The
Philodendron Dilcher and Daghlian (1977)
Araliaceae temperate plants were mostly part of the upland vegetation
Dendropanax (as Aralia, Dilcher and Dolph (1970) of the southern Appalachian Mountains. The Fagaceae
Berry, 1916) (probably Castaneoideae, Quercoideae, and the extinct Dry-
Ceratophyllaceae ophyllum) are abundant throughout the Middle Eocene,
Ceratophyllum Herendeen et al. (1990)
while Quercus pollen especially characterizes the latest
Euphorbiaceae
Crepetocarpon Dilcher and Manchester (1988) Claiborne-Vicksburg vegetation and was likely abundant
Hippomaneoidea Crepet and Daghlian (1982) on the Gulf Coastal Plain (Frederiksen, 1981). Like many
Fagaceae taxa in North American Tertiary floras, Eucommia pres-
Castaneoidea Crepet and Daghlian (1980) ently grows in eastern Asia (one species, E. ulmoides, in
Juglandaceae
Crepet et al. (1975)
China), but it was widespread during the Eocene and Early
Eokachyra
Eoengelhardia Crepet et al. (1980) Oligocene. Megafossils are known from the Beaverhead
Oreoroa Dilcher and Manchester (1986) Basins, Claiborne, Clarno, Florissant, Green River, John
Pamoreom unnea Dilcher et al. (1976) Day, Klondike Mountain (Republic), and other floras (Call
Lauraceae and Dilcher, see also Magallon-Puebla and Cevallos-Ferriz,
Ocotea Dilcher (1963)
1994). The stratigraphic range is from late Early Eocene to
Leguminosae
Caesalpinioideae Early Oligocene in North America (north of Mexico). Mi-
Crudia Herendeen and Dilcher (1990a) crofossils range from the Late Paleocene (Powder River
Caesalpinia Herendeen and Dilcher (1991) Basin, Wyoming) to the Oligocene.
Mimosoideae A summary of the principal palynomorphs from the
Eomimosoidea Crepet and Dilcher (1977)
Daghlian et al. (1980)
succeeding Jacksonian provincial stage is given in Table
Duckeophyllum Herendeen and Dilcher (1990b) 6.3. The nature of the vegetation is difficult to reconstruct
Eliasofructus Herendeen and Dilcher (1990b) from palynological evidence because a number of extinct
Parvileguminophyllum Herendeen and Dilcher (1990b) or unknown forms are present, many are of a generalized
Papilionoideae morphology and can be referred only provisionally to clus-
Diplotropis Herendeen and Dilcher (1990c)
ters of genera or families, and tropical pollen is compara-
Nyssaceae
Nyssa Dilcher and McQuade (1967) tively poorly known (Frederiksen, 1981). By the Late
Oleaceae Eocene a significant number of fossil botanical taxa are
Fraxinus Call and Dilcher (1992) similar morphologically to modern forms and can be
Ulmaceae sorted into ecologicaly consistent and meaningful assem-
Eoceltis Zavada and Crepet (1981)
blages. This is in contrast to the impression from some lit-

erature on Tertiary and Quaternary paleoecology that eco-
logical requirements and generic composition of all but the
most modern of floras have so shifted that the modern ana-
log method has virtually no value. This has not been my
experience and I do not believe it is true. The Middle
Eocene, however, is a transition period in the moderniza-
tion of floras and lineages. Consequently, there is a range in
the confidence level of biological identifications expressed
in the nomenclature from relatively certain (e.g., Nipd], to
probable (cf. Carya], to possible [Clethrai], to merely desig-
nating a comparable morphotype to convey information on
pollen and spore structure (Eugenia/Myrtusl, Fagus-type).
If a specimen is not exactly comparable to a modern ana-
log, it may later be found to represent a completely unre-
lated taxon. These annotations are important in evaluating
the taxonomic and biogeographic significance of the pref-
aced names.
In the absence of Avicennia, Laguncularia, Rhizophora,
and other New World mangroves, coastal brackish-water
habitats were occupied by Mpa, which was possibly asso-
ciated with Acrostichum that presently occurs with man-
groves. Nipa was widespread in the northern hemisphere
during the Eocene (Tralau, 1964), but it became extinct
there and in the Caribbean region and Europe, near the
Eocene-Oligocene boundary, and occurs today only in
tropical southeast Asia. On sandy flats, but removed from
the tidal influence, Pinus and an understory of Sabal-
Serenoa-type palms and Ephedra were present (early sand
pine scrub association). Nyssa, members of the Taxodi-
aceae complex, and probably Carya grew in and marginal
Middle Eocene through Early Miocene History 195
Figure 6.3. Generalized paleovegetation
diagram for North America during the late
Middle Eocene. C, mixed coniferous for-
est; D, polar broad-leaved deciduous for-
est; N, notophyllous broad-leaved forest;
P, paratropical rain forest; S, tropical semi-
deciduous forest; T, tropical rain forest.
Reprinted from Upchurch and Wolfe
(1987) with the permission of
Cambridge University Press.
to freshwater swamps (flood-plain forest association). Most
of the inland, well-drained, low to midelevation landscape
apparently was covered by a mosaic of paratropical and
warm-temperate genera. Paratropical representatives in-
clude Anemia-Mohria, Lygodium, various Palmae, and
possibly some Anacardiaceae, Bignoniaceae, Bomba-
caceae, Loranthaceae, Rutaceae, Sapindaceae, and Sapo-
taceae that cannot be identified to genus. Jacksonian-age
pollen of genera that presently range into warm-temperate
zones includes Podocarpus, Acer?, Alnus, various Faga-
ceae, Ilex, Platanus, Pterocarya, and Ulmus-type. At the end
of the Eocene near the Jackson (Upper Eocene)-Vicksburg
(Lower Oligocene) boundary (34-35 Ma, calcareous nano-
fossil zone NP21), pollen of Dryophyllum-Quercus (oak)
increases greatly as a result of a cooler and drier climate
compared to Wilcox and Claiborne time (Frederiksen, 1981).
Many of the plants probably ranged into cool-temperate up-
land habitats where Picea and Tsuga were present. This
warm-temperate to subtropical forest is reminiscent of the
drier phase of present-day vegetation along the lower east-
ern slopes of the Sierra Madre Oriental in Mexico between
1000 and 2000 m elevation where Podocarpus, Acer, Al-
faroa, Alnus, Carya, Celtis, Ilex, Juglans, Myrica, Platanus,
Tilia, Ulmus, and others are associated with tree ferns,
Clethra, Ficus and other Moraceae, various Myrtaceae,
Leguminosae, and other tropical trees and shrubs. Man-
groves fringe the coast there, and cool-temperate plants
grow in the adjacent highlands [presently Abies and
species of Pinus (e.g., P. montezumae, P. rudis) and Picea
until recent times]. Huatusco is located in this zone at an
Table 6.3. Palynomorphs from Jacksonian provincial stage (Upper middle and Upper Eocene, -40—37 Ma) of Gulf Coast
United States. After Frederiksen, 1980a, 1981.
Palynomorphs Affinities

Bryophytes
Sphagnum Sphagnaceae

Pteridophytes
Lycopodium Lycopodiaceae
Selaginella Selaginellaceae
Acrostichum'? Pteridaceae
Cicatricosisporites dorogensis Anemia-Mohria, Schizaeaceae
Laevigatosporites haardtii Various Filicales (ferns)
Lygodiumsporites adriennis Schizaeaceae, Lygodium
Verrucatosporites alienus Various Filicales (ferns)

Gymnosperms
Cedrus Pinaceae
Cryptomeria-Metasequoia-Sequoia Taxodiaceae
Ephedra Ephedraceae
Picea Pinaceae
Pinus tenuextima Pinaceae
Podocarpus Podocarpaceae
Taxodiaceae - Cupressaceae -Taxaceae
Tsuga Pinaceae

Monocotyledon Angiosperms
Aglaoreidia pristina Unknown
Arecipites pseudotranquillus Palmae
Liliacidites tritus Pseudophoenixl, Palmae
Milfordia sp. Prob. Restionaceae
Monocolpopollenites tranquillus Palmae
Nipa sp. Palmae
Proxapertites sp. Unknown, Araceae or Annonaceae?
Pseudophoenix"? sp. Palmae
Sabal Palmae
Serenoctf Palmae

Dicotyledon Angiosperms
Acer'? Aceraceae
Ailanthipites berryi Unknown
Alangium"? Alangiaceae
Albertipollenites"? araneosus Bignoniaceae
Alnus Betulaceae
Anacolosa-Cathedra-Ptychopetalum Olacaceae
Bumelia"? Sapotaceae
Caprifoliipites sp. Viburnum"?, Caprifoliaceae
Carya Juglandaceae
Celtis Ulmaceae
Cupanieidites orthoteichus Loranthust, Loranthaceae
Cupania"?, Sapindaceae
Cupuliferoidaepollenites liblarensis Dryophyllum"?, Fagaceae
C. certus Cassia"?, Leguminosae
C. brevisulcatus Chrysophyllum"?, Sapotaceae
CyrillaceaepoUenites megaexactus Cyrilla-Cliftonia
Diospyros"? Ebenaceae
Fraxinoipollenites variabilis Unknown
F. sp. Unknown
Intratriporopollenites sp. Tiliaceae-Bombacaceae
continued
 Middle Eocene through Early Miocene History 197

Table 6.3. continued

Juglans Juglandaceae
Ludwigia Onagraceae
Manilkara? Sapotaceae
Momipites coryloides Juglandaceae, Engelhardia group
M. microfoveolatus Juglandaceae, Engelhardia group
Myrica propria Myricaceae
Myriophyllym Haloragidaceae
Myrtaceidites sp. Eugenia—Myrtusl, Myrtaceae
Nuxpollenites Sapindaceae?
Nyssa Nyssaceae
Parthenocissusl Vitaceae
Platanus Platanaceae
Pollenites modicus Unknown
P. pseudocingulum granulatus Fagaceae?
Pseudolaesopollis ventosa Costaea?, Cyrillaceae
Pterocarya Juglandaceae
Quercoidites microhenricii Fagaceae
Rhoipites Rhust (Anacardiaceae), Cornaceae, Nyssaceae
Salixipolllenites sp. Fraxinusi, Oleaceae
Siltaria sp. (<25um) Fagaceae
Striopollenites terasmaei Unknown
Symplocos Symplocaceae
Tetmcolporopollenites sp. Sapotaceae
Tilia Tiliaceae
Tricolporopollenites illiacus Ilex, Aquifoliaceae
Ulmus-Planera-Zelkova type Ulmaceae
Verrutricolporites sp. Unknown
Zanthoxyluntf Rutaceae
Adapted from Frederiksen (1980a, 1981).

elevation of 1344 m. There the MAT is 15.7°C, the mean
monthly range is 7°C [between 15°C (December-January)
and 22°C (May)], and the annual rainfall is 1745 mm. The
MAT extrapolated downward 1 km to 344 m is 18.7°C
using a lapse rate of 3°C /km (Wolfe, 1992b) or 21.6°C at a
lapse rate of 5.9°C/km (Meyer, 1992). Sedimentological ev-
idence also indicates that from the Late Eocene to the
Miocene the rainfall in Texas was gradually decreasing and
becoming more seasonal (winter dry; Folk, 1955).
The vegetation of the Mississippi Embayment region, or
any other area of North America, was not static even at the
time intervals of stages. Frederiksen (1991) has shown that
the general deterioration of terrestrial climate between the
Middle Eocene (Lutetian, Bartonian) to earliest Oligocene
(Rupelian) encompassed four events of rapid vegetation
change. One was a turnover at the beginning of the Luetian
(early Middle Eocene) as reflected by peaks in pollen of
first occurrences (Platanus, Cyrilla-Cliftonia, large grains
of Carya) and last occurrences (mostly Paleocene species;
Jarzenipollis trina, PistillipoHenites macgregorii, Momip-
ites tenuipolus group). The second event, occurring be-
tween the latest Lutetian and early Bartonian, has been
termed the Middle Eocene Diversity Decline (MEDD) by
Frederiksen (1991; Cook Mountain Formation, Texas-
Louisiana; Lisbon Formation, Alabama; P13-P14, NP16).
Some new taxa appeared, including the first grasses, and
Pterocarya and Ephedra migrated in from the west. Over-
all, however, the elimination of many Paleocene and Early
Eocene arborescent taxa in the Palmae, Juglandaceae, Ul-
maceae, and probable Eucommiaceae resulted in a net de-
crease in diversity as a result of reduced and more seasonal
rainfall. The third event was an increase in pollen similar
to modern Quercus in the upper Priabonian (Late Eocene),
reflecting the temperature decline and redistribution of
rainfall. The continuation and intensification of this trend
eventually produced the fourth event, which was a further
reduction in diversity as more tropical elements were elim-
inated from the vegetation of the southeastern United
States (the Terminal Eocene Event; now Early Oligocene;
NP21). Evidence of these stepwise changes in climate, and
the paralleled response by the biota, is preserved in the pal-
ynological and megafossil records (Wolfe, 1978; Wolfe and
Poore, 1982), but they are out of phase and occurred later
in the megafossil analyses. The differences between the two
records are further discussed by Frederiksen (1988), but the
reason for the disparity is not known. There were also local
differences in vegetation and other changes that occurred
over comparatively short periods of time. For example,
swamp vegetation in the Claibornian is dominated by Fa-
gaceae, while in the Jacksonian the prominent pollen types
in swamp deposits are Juglandaceae (Momipites).
The purpose of this excursion into detailed changes in
198 Late Cretaceous and Cenozoic History of North American Vegetation
climate and communities is to avoid the misconception
that vegetational history followed an even course as might
be suggested by the generalized benthic paleotemperature
curve (Fig. 3.1). A similar impression might be created by
the emphasis in this text on major trends that lead to the
modernization of North American vegetation at the forma-
tion and association levels. Significant alterations in the
environment were also occurring on a much finer time
scale, including the Milankovitch variations, which were
periodically overprinted or enhanced by such factors as
continentality, variations in CO2 concentration, alterations
and new threshold levels in ocean circulation, but which
are less apparent at the resolution of tens of thousands to
millions of years. As noted earlier in Chapter 5, Wing
(1996) has shown that in the northern Rocky Mountains a
previously unrecognized cool interval of 1—1.5 m.y. dura-
tion occurred at the Paleocene—Eocene boundary. This was
within an overall trend toward the Early Eocene thermal
maximum. When a sufficient number of floras are carefully
studied from a region to give high resolution, trends can
often be seen that are obscured by general collecting. New
data are emerging that suggest abrupt climatic changes su-
perimposed on long-term trends are the rule. The pace
probably did increase in the Late Cenozoic, but the fre-
quency and intensity of the changes are also obscured in
older sediments.
It is now recognized that these fluctuations often corre-
late with rapid peaks in evolution (Rea et al., 1990). A sharp
decrease in 18O in seafloor sediments off Antarctica indi-
cates that during the seemingly gradual warming at 55 Ma,
deep-sea temperatures rose from 10 to 18°Cin 2 Ky (Kennett
and Stott, 1991). This not only caused an extinction of 40%
of the benthic foraminifera, but the event correlates with a
rapid change from archaic terrestrial mammals to modern
forms in the Bighorn Basin of Wyoming and Montana in the
Early Eocene. The correlation between these two geograph-
ically distant events is preserved in carbon isotope patterns.
The temperature rise caused a release of CO2 from the ocean
surface to the atmosphere, which was incorporated into the
tissue of land plants via photosynthesis and eventually
made its way into the teeth and bone of grazing and brows-
ing land animals. When carbon isotopes were measured
from the rapidly evolving Bighorn Basin herbivores, the
temperature spike was found at the same time as the 18O de-
tected increase in the latest Paleocene-Early Eocene tem-
perature gradient from Antarctica. Koch et al. (1992) report
that the 12C/13C ratio in enamel apatite and paleosol carbon-
ate decreased by 5.9 and 4.5%, respectively (increasing
temperatures) in less than 50 Ky and persisted for less than
100 Ky. An appreciation of the nature of biotic history war-
rants an occasional reminder that stasis was no more a char-
acteristic of past times than it is of the present, that what
appears as a gradual trend on a broad scale is actually punc-
tuated by rapid shifts in climate, and that these shifts are
now being recognized as forcing mechanisms for plant and
animal evolution (Bennett, 1990).
The same general sequence of Late Eocene to Early
Oligocene vegetational change described for the Gulf Coast
is also evident in the Mays Landing core from southeastern
New Jersey. By the Late Oligocene common palynomorphs
there included Pinus, Podocarpus, Tsuga, Taxodiaceae-
Cupressaceae type, Alnus, Betula, Carya, Fagus, Ilex, Liq-
uidambar, Momipites (Alfaroa-Engelhardia-Oreomunnea),
Nyssa, Quercus (Owens et al., 1988), and other dominants
of the modern deciduous forest formation.
The climatic sequence suggested by the plant microfos-
sil record (Frederiksen, 1991; Owens et al., 1988) is similar
but not identical to that from the marine realm (Prothero,
1989; Table 6.4). Further studies may eventually bring the
patterns into closer accord, allowing for lag time and dif-
ferences in the nature and degree of forcing mechanisms
between marine and terrestrial environments. In the mean-
time, it is clear that temperatures during the Middle
Eocene to Middle Oligocene downshifted in stages, in part
due to decreases in CO2 concentration and the initiation of
glaciation on Antarctica, which cooled marine waters (less
evaporation) and the atmosphere (less available moisture),
and altered ocean and atmosphere circulation.
As a generalization, the Late Eocene and later vegetation
of the southeastern United States records the gradual elim-
ination of many tropical species; the reorganization, expan-
sion, and diversification of temperate species in response to
cooling; a widening between cold—warm monthly aver-
ages; and reduced and greater seasonality in rainfall. The
Middle Eocene through the earliest Oligocene is the transi-
tion phase between tropical and temperate conditions in
the midnorthern latitudes, and the Jacksonian and related
microfossil assemblages reflect this period of change in cli-
mate and vegetation.
Quantitative estimates of the Middle Eocene climate for
the southeastern United States, dating back to the time of
Berry in the early 1900s, range widely and include impre-
cisely defined and overlapping terms noted earlier. More
recently, the vegetation has been characterized as a tropical
dry forest with a MAT fluctuating between 20 and 28°C
(Wolfe, 1978; 22 and 30°C, Wolfe and Poore, 1982). Dilcher
(1973) suggests a climate like that of southern peninsular
Florida or coastal Louisiana (MAT 24°C, 21°C, respec-
tively), with occasional frosts in the highlands, as does Ax-
elrod (1966a) and Frederiksen (1980b; see previous Hua-
tusco data extrapolated to sea level; 18.7-21.6°C ). The
tropical dry forest was eliminated in the southeastern
United States after the Eocene by cooling temperatures and
as changes in atmospheric circulation, induced by contin-
ued uplift of the Rocky Mountains, began bringing greater
precipitation and warm moist winds from the south.
To the far northeast on Ellesmere Island (latitude 81° 30'
N), sediments correlated with the Eocene portion of the
Late Cretaceous to Eocene Eureka Sound Group have
yielded a rich mammalian fauna, along with large sala-
manders, anguid and varamid lizards, a boid snake, tur-
tles, tortoises, and alligator (Estes and Hutchison, 1980;
 Middle Eocene through Early Miocene History 199

Table 6.4. Relationship between marine, Mississippi Embayment, and Mays Landing events, documenting stepwise
nature of Middle Eocene to Middle Oligocene climatic change and vegetation response.
Mississippi
Marine3 Embaymentb Mays Landingc
Middle Eocene 1st pollen event Early Eocene-
Lutetian -B artonian Early Middle Eocene Middle Eocene
-41 Ma 2nd pollen event boundary
Middle Eocene (MEDD) Middle Eocene
Middle Eocene-Late Middle Eocene-
Eocene boundary Late Eocene
Bartonian-Priabonian boundary
-37 Ma
Late Eocene 3rd pollen event
Late Priabonian Late Eocene
-35 Ma
Eocene-Oligocene
boundary
33.5 Ma 4th pollen event
Early Oligocene
Middle Oligocene
Late Rupelian Middle Oligocene
-30.5 Ma 31-29 Ma
"Data from Prothero (1989).
b
Data from Frederiksen (1991).
°Data from Owens et al. (1988).

McKenna, 1980). A pollen flora from Lake Hazen, northern
Ellesmere Island (MacGregor, in Christe, 1964; Christe and
Rouse, 1976) includes Cedrus, Picea, Pinus, Tsuga, Acer,
Betula, Carya, Corylus-Carpinus, Castanea, Fagus, Ptero-
carya, and Quercus. On Axel Heiberg Island, the Buchanan
Lake Formation preserves fossil floras from a floodplain en-
vironment. The palynoflora adds Larix, Metasequoia,
Alnus, Diervilla, Engelhardia, Fraxinus, Juglans, Lonicera,
Tilia, and Viburnum to the Ellesmere assemblage (Mcln-
tyre, 1991). The vegetation was a temperate mixed hard-
wood—conifer forest similar to the modern lake states
forest and Appalachian highlands to the south. An adja-
cent megafossil flora preserves holdovers from the warm-
temperate broad-leaved deciduous forest that had occu-
pied these latitudes in the Early Paleocene (Fig. 5.12),
including Larix, Pinus, Taxodium, Acer, Betula, Cercidi-
phyllum, Corylus, Platanus, Populus, and Ulmus. Records
of Cruciptera, Paleocarya, and Hooleya (extinct genera of
the Juglandaceae) in the Tertiary of western North America,
in the Middle Eocene pipe clays of southern England, and
in Messel, Germany, document that the North Atlantic land
bridge was fully functioning (Manchester et al,, 1994).
Cold-month temperatures were 0-4°C (generally above
freezing), warm-month temperatures ~25°C, and MAT is
estimated at 12-13°C (Axelrod, 1984; 12-10°C; Basinger et
al., 1994). McKenna (1983) places coastal water tempera-
ture bordering Ellesmere Island in the Middle Eocene at
15°C.
A MAT of 24°C for the southeastern United States at 35°
N and 13°C for Ellesmere Island at 81° N calculates to a
Middle Eocene gradient of 0.28°C /1° latitude. This com-
pares with 0.3°C /1° latitude in the Maastrichtian, indicat-
ing that by the Middle Eocene the MAT had declined
episodically from Early Eocene maxima to about what they
were at the end of the Cretaceous (Fig. 3.1). The trend con-
tinued through Middle and Late Cenozoic time to the pres-
ent gradient of ~0.5°C /1° latitude.
Little is known of Late Eocene vegetation of the Plains
region, although woods of Robinia, Prunus, and Zelkova
are known from the Chadronian of Nebraska and indicate a
seasonal climate (Wheeler and Landon, 1992). Paleosols,
root diameters, and tree spacing in the Badlands of South
Dakota suggest rainfall decreased from 1000 mm (38 Ma,
Late Eocene) to 250-450 mm (29.5 Ma, Late Oligocene).
The shift in vegetation inferred from this sequence is from
moist forest (38 Ma), to dry forest (34 Ma), to dry woodland
(33 Ma), to wooded grassland with gallery forests (32 Ma),
to more open savanna with an increasing understory of
grasses by 30 Ma. Land snails show a similar shift at the
Eocene-Oligocene boundary from large-shelled subtropi-
cal forms (MAT of 16°C and annual precipitation of 450
mm) to the smaller warm-temperate ones presently occu-
pying open woodlands with a pronounced dry season
(Evanoff et al., 1992). In the High Plains and northern
Rocky Mountains, Early Eocene megathermal tropical and
paratropical vegetation, which had extended north to
60°-65° N, started to be replaced by microthermal decidu-
ous communities.
200 Late Cretaceous and Cenozoic History of North American Vegetation
In the central Rocky Mountains crustal deformation
slowed, but volcanism deposited ash up to 1700 m thick in
the basins. Elevations had reached a point where vegeta-
tion east and west of the continental divide was develop-
ing differently (Leopold and MacGinitie, 1972; Wing, 1987;
Fig. 5.8, hatched area). To the east in the earliest part of the
Middle Eocene, a seasonally dry, subtropical open wood-
land-savanna phase is well documented (34.8% in the
playa lake Little Mountain flora of southeastern Wyoming,
51-50 Ma; Leopold and MacGinitie, 1972) and paleotropi-
cal elements constituted 13%, declining from 16.7% in the
Wind River flora. Members included Alchornea, Cardiosper-
mum, Populus, and microphyllous leguminosae. Slightly
later in the early Middle Eocene, tropical and subtropical
components reached a peak of -50% in the Tipperary and
Kisinger lakes flora of the Aycross Formation in the Wind
River Basin of northeastern Wyoming (MacGinitie, 1974;
-49-50 Ma) due to an increase in neotropical (15.5%) and
pantropical (22.0%) species. Non-entire-margined leaves
reached just over 50%, which signifies a subtropical forest
(wet summers, dry winters) in the leaf physiognomy sys-
tem, and about 60% were deciduous. Members included
Acrostichum, Thelypteris, Chamaecyparis, Glyptostrobus,
Apeiba, Canavalia, and Dendropanax. In the middle Mid-
dle Eocene (Green River time, 45-48 Ma) tropical and sub-
tropical species declined to 33.8%, and New World taxa
constituted 23.4% of the flora.
The change was gradual and complex from an Early
Eocene and older tropical-paratropical vegetation with
primarily Asian affinities in the central Rocky Mountains
and Plains to a seasonally dry subtropical Middle Eocene
vegetation with Mexican-Central American affinities. Some
seasonally dry subtropical elements from the present-day
Latin American flora actually began to appear in the Pale-
ocene-Eocene (Clarkforkian, Wasatchian). Examples in-
clude Chaetoptela microphylla, Woodwardia gravida, and
Populus sect. Abaso (Wing, 1987). Some Asian elements
were already locally extinct in the eastern Rocky Moun-
tains, including Metasequoia and members of the Cercidi-
phyllum complex. Other records that are exceptions to the
overall trend are the Asian Platycarya and members of the
Flacourtiaceae, Icacinaceae, and Menispermaceae, which
were present by the Wasatchian and continued through the
Bridgerian. Ailanthus is recorded in the Chalk Bluffs and
Green River floras (Early to Middle Eocene) and persists
through the Late Eocene-Oligocene (Ruby and Beaverhead
Basins floras of southwestern Montana; Becker, 1961,1969)
and into later Tertiary time. Nonetheless, the trend from
primarily Old World to New World tropical-subtropical
affinities is evident among the floras. Cooling temperatures
that reduced migration through Beringia was a factor.
The Green River flora and the Thunder Mountain flora
are of approximately the same age, but were deposited at
different elevations on the east and west side of the Rocky
Mountains, and give some insight into the effect of the de-
veloping continental divide and the zonation of western
American vegetation during the Middle Eocene. The Green
River flora (-45 Ma, Roehler, 1992a,b; 45-48 Ma, Wing and
Greenwood, 1993) of northwestern Colorado and north-
eastern Utah is preserved in lacustrine (lake) sediments.
The formation covers an area of -65,000 km2 (25,000 mi2);
and some of the lake sediments are varved, reflecting sea-
sonal deposition. The sediments were laid down near the
end of the Laramide orogeny at ~pl 35° N (5°-8° south of
its present location; Roehler, 1993), placing it near the
juncture of the Hadley and Rossby circulations where high
pressures dominated and winds were primarily from the
west. Surrounding mountains were 1-1.3 km higher than
the basin, and rainshadows likely contributed to the dry-
ness of the intermontane basin. Seventy-seven deposi-
tional cycles are present in the basin and they appear to
conform to the 23 Ky precession and the 100 Ky eccentric-
ity cycles. The Green River Formation is similar in age to
the Claiborne Formation of the Mississippi Embayment re-
gion, indicating that the seasonally dry subtropical climate
of the Gulf region extended to the central Rocky Mountains
during the Middle Eocene.
The fossils were deposited in three principal basins oc-
cupied by great shallow lakes: Uinta Basin (Lake Uinta),
Green River Basin (Lake Gosiute), and Fossil Basin (Fossil
Lake; Figs. 6.4, 6.5). The Green River Formation is known
for its exceptionally well-preserved fish fauna and for the
greatest reserves of oil shale presently known (carbonate
rocks rich in kerogen). Genera reported by MacGinitie
(1969) are given in Table 6.5. The most abundant speci-
mens are ofLygodium (abundant), palm leaves (abundant),
Typha (common), Musophyllum (common), Zingiberopsis
(common), Canavalia (common), and Sapindus (abun-
dant). Ceratophyllum (Herendeen et al., 1990) and Populus
(Fig. 6.6A; Manchester et al., 1986) have been confirmed,
and Caesalpinia (Herendeen and Dilcher, 1991) and Eu-
commia (Call and Dilcher, 1997) have been added. Other
members include Acer (Fig. 6.6C), Engelhardia (Fig. 6.6B),
Platanus (Fig. 6.6D), Pterocarya (Fig. 6.6F), and Zelkova
(Fig. 6.6E). Pollen (mostly Wodehouse, 1932, 1933) in-
cludes Ephedra, Picea, Alnus, Betula, Carya, Nyssa, Platy-
carya, and Tilia. The vegetation occupied four habitats: lake
margins, swamps, and floodplains at 300 m elevation (Pla-
tanus, Populus, Salix); drier, well-drained, bordering
slopes up to 700 m above the lake level (Ephedra, some
Pinus species, Cardiospermum, Celtis, Quercus, Sapindus,
Vauquelinia; savanna, oak-pinon pine woodland); up-
lands 1300 m above the lake level (mostly broad-leaved de-
ciduous trees; Acer, Carya, Engelhardia, Gymnocladus,
Liquidambar, Liriodendron); and a mixed hardwood-
coniferous forest zone at 1000 m and extending possibly to
1700 m (Picea, other Pinus species, Betula). Rainfall is esti-
mated to have been -700 mm (28 in.; subhumid) and sea-
sonal.
MacGinitie (1969) interpreted the slope vegetation dur-
ing Green River time as a unique type of savanna wood-
land with widely spaced trees and shrubs but with little or
 Figure 6.4. Sedimentary deposits of Eocene Lakes Uinta
(Uinta and Piceance Creek Basins) and Gosiute (Green
River and Washakie Basins) and Fossil Lake (Fossil Basin).
(1) Green River Formation; (2) bedded salts; (3) basement.
Reprinted from Eugster (1982; based on Eugster and
Hardie, 1978) with the permission of the Geophysics
Study Committee (1982) of the National Academy of Sci-
ences. Courtesy the National Academy Press.

Figure 6.5. (A) Carr Creek Canyon from Wardell Ranch, Green River flora, northwestern Colorado. (B) Fossil plant local-
ity southeast of Bonanza, Utah. Evacuation Creek Member, Green River flora. Repinted from MacGinitie (1969) with the
permission of the University of California Press.
Table 6.5. Megafossils from late Middle Eocene Green River flora of northwestern Colorado and northeastern Utah.

Pteridophytes
Aspleniaceae Schizaeaceae
Asplenium delicatula Lygodium kaulfussii
A. sermforme
Equisetaceae Salviniaceae
Equisetum Winchester! Azolla berryi
Isoetaceae Pteridaceae
Isoetites horridus Acrostichum hesperium

Gymnosperms
Pinaceae Taxodiaceae
Pinus balli Sequoia cf. affinis
P. florissanti

Monocotyledon Angiosperms
Potamogetonaceae Typhaceae
Potamogeton rubus Typha lesquereuxi
Sparganiaceae
Sparganium antiquum
S. eocenicum

Dicotyledon Angiosperms
Aceraceae Lauraceae Rosaceae
Acer lesquereuxi Beilschmiedia eocenica Prunus stewarti
Dipteronia insignis Lindera allardi Rosa hilliae
Anacardiaceae Ocotea coloradensis Vauquelinia comptonifolia
Anacardites schinoloxus Persea coriacea Rutaceae
Astronium truncatum Leguminosae Ptelea cassioides
Rhus nigricans Caesalpinites falcata Salicaceae
Toxicodendron Winchester! C. pecorae Populus cinnamomoides
Apocynaceae Erythrina roanensis P. wilmattae
Apocynospermum coloradensis Gymnocladus hesperia Salix cockerelli
Araliaceae Leguminosites lesquereuxiana S. longiacuminata
Araliophyllum quina L. regularis Sapindaceae
Oreopanax elongatum Mimosites coloradensis Allophylus flexifolia
Aristolochiaceae ParvileguminophyUum coloradensis Athyana balli
Aristolochia mortua Swartzia wardelli Cardiospermum coloradensis
Berberidaceae Malvaceae Koelreuteria viridifluminis
Mahonia eocenica Hibiscus roanensis Sapindus dentoni
Bombacaceae Meliaceae Thouinia eocenica
Ochroma murata Cedrela trainii Simarubaceae
Burseraceae Menispermaceae Ailanthus lesquereuxi
Bursera inaequalateralis Menispermites limaciodes Sterculiaceae?
Celastraceae Myrtaceae Sterculia coloradensis
Celastrus Winchester! Eugenia americana Styracaceae
Euphorbiaceae Oleaceae Styrax transversa
Aleurites glandulosa Osmanthus praemissa Symplocaceae
Fagaceae Platanaceae Symplocos exilis
Quercus cuneatus Platanus wyomingensis Tiliaceae
Q. petros Proteaceae Triumfetta ovata
Hammamelicaceae Lomatia lineatulus Ulmaceae
Distylium eocenica Rhamnaceae Celtis mccoshii
Liquidambar callarche Berchemiopsis paucidentata Zelkova nervosa
Juglandaceae
Engelhardtia uintaensis
Pterocarya roanensis

Adapted from MacGinitie (1969). The specimen reported as Fraxinus flexifolia (Brown, 1940) is now considered to be a mimosoid leaflet
Parvileguminophyllum coloradensis (Call and Dilcher, 1994).

202
Figure 6.6. (A) Populus wilmattae, Green River flora. (B) Engelhardia uintaensis, Green River flora. (C)
Acer lesquereuxi, Green River flora. (D) Platanus wyomingensis, Green River flora. (F) Pterocarya roa-
nensis, Green River flora. (E) Zelkova nervosa, Green River flora. Reprinted from MacGinitie (1969) with
the permission of the University of California Press.
204 Late Cretaceous and Cenozoic History of North American Vegetation
no grass. He called it Orizaban-subtropical, by analogy
with seasonally dry woodland savanna on Mt. Orizaba,
Mexico. Wolfe (1985) believes it was a version of the semi-
deciduous tropical dry forest that characterized the south-
eastern United States during the Middle and Late Eocene.
Independent evidence from sedimentology confirms the
subhumid aspect of the climate (Eugster, 1982). Evapora-
tion exceeded inflow most of the time, and at least 25 epi-
sodes are recorded where Lake Gosiute dried up com-
pletely and became a salt pan. In general, the Green River
differs from the older tropical Wind River flora with dis-
tinct Asian affinities in being dry subtropical with greater
Latin American affinities. Varves in the lake sediments in-
dicate that long warm dry summers alternated with cool
moist winters.
It has been suggested that the meager representation of
grasses in the Eocene and Oligocene fossil record may not
be solely due to their rarity in the vegetation. Thomasson
(1985) believes the later Miocene record is too diverse to ex-
clude their presence at least by the Oligocene and that their
appearance marks the development of preservable silicified
tissue. However, fossil pollen is also rare, and until some
tangible evidence for a grass understory is found, the vege-
tation can best be described as a woodland-savanna or trop-
ical dry (oak-pinon pine) forest or woodland.
Early computer models for Eocene climates of the con-
tinental interior yielded simulations that were in conflict
with results from paleontology. The CCM and energy bal-
ance model hindcasted winters with temperatures of
-30°C (Barren, 1983; Sloan and Barron, 1990), while a
mass of biological and geological information clearly indi-
cated minimum winter temperatures at or above freezing.
However, the models did not incorporate the boundary
condition of large lakes. When the Green River lakes were
added to the simulations, with a CO2 concentration twice
that of the present, the difference between the model and
ground data was largely resolved. In fact, the model re-
sponded as strongly to the presence of the lakes as to the
level of CO2 (Sloan, 1994).
A flora from the Germer Member of the Challis Vol-
canics (47-46 Ma) was deposited at an elevation of-600 m
and illustrates the difference in vegetation east and west of
the continental divide. While the Green River flora to the
east at lower elevations was a woodland-savanna or trop-
ical dry oak-pinon pine woodland, the Germer flora repre-
sents a mixed deciduous hardwood-coniferous forest of
Aesculus, Alnus, Betula, Carya, Cercidiphyllum, Eucom-
mia, Juglans, and Ostrya, associated with frequent to abun-
dant Metasequoia, Pseudolarix, and Sequoia and lesser
amounts of Abies, Chamaecyparis, Picea, Pinus, and Tsuga
(Axelrod et al., 1991). This ecotonal forest type was com-
mon at the transition between the deciduous forest and the
western montane coniferous forest in the Middle Tertiary
when broad-leaved deciduous angiosperms were more
prominent. It now has been mostly eliminated as the de-
ciduous forest in the west declined with the drying condi-
tions associated with uplift of the coastal mountains.
Clearly by the Middle Eocene the Rocky Mountains were
directly influencing the course of vegetational history by
atmospheric circulation and rainshadow mechanisms,
while the coastal mountains were not yet interfering with
moisture transport from the Pacific Ocean.
The Thunder Mountain flora of Middle Eocene age (-45
Ma; Axelrod, 1990) in central Idaho, 90 km northwest of
the Germer flora, was deposited at an elevation of -1550 m.
It illustrates some of the elevational zonation evident in
floras on the west side of the Rocky Mountains. The Road
Florule is lowest in elevation and includes Abies, Larix,
Picea, Pinus, Tsuga, Sequoia, Chamaecyparis, and Thuja,
along with Alnus, Amelanchier, Arctostaphylous, Cas-
tanea, Comptonia, Cornus, Mahonia, Malus, Populus, Pax-
istima, Pterocarya, Quercus, Rhamnus, Rhododendron,
Salix, and Viburnum. It represents the upper elevational
zone of a mixed hardwood-coniferous forest. The Dewey
Florule is a higher elevation assemblage that descended to
lower elevations along streams and canyons through cold-
air drainage. It includes Abies, Larix, Picea, and Pinus, but
only a few deciduous hardwood species, and is a western
montane coniferous forest. The MAT is estimated at 8.5°C,
tree line at 2195 m, and precipitation between 1100 and
1150 mm (mostly summer rain and light winter snow).
The presence of Abies and Picea reveals that elements
of the western montane coniferous forest were already es-
tablished in western North America by the Middle Eocene.
These began to coalesce into the modern formation as ele-
vations increased and temperatures declined. The pres-
ence of Ephedra, Celtis, and Ocotea in the Green River
flora further documents that plants adapted to subhumid
habitats were also available for later assemblage into desert
and other dry to arid vegetation with the rise of the Coast
Ranges, Sierra Nevada, and Cascade Mountains, princi-
pally in the Pliocene. Vegetational history is a record of
once-isolated preadapted elements or restricted early ver-
sions of a vegetation type present in edaphically or physio-
graphically controlled sites, responding to environmental
change by assemblage, diversification, and expansion into
new kinds of communities. This conceptual view of the
dynamics of community evolution is nicely presented by
Axelrod (1992) with regard to the Eocene Elko flora (36 Ma)
of northeastern Nevada and three Miocene floras (15-16
Ma) from western Nevada (Middlegate, Eastgate, and Buf-
falo Canyon; Wolfe et al. 1997, Table 1).
Slightly younger vegetation to the east of the central
Rocky Mountains is preserved in lacustrine shales of the
latest Eocene to earliest Oligocene Florissant Formation of
central Colorado (Fig. 6.7, Table 6.6) dated at -35 Ma (Epis
and Chapin, 1974; adjusted to new decay constant). West-
ern North America rotated southward during the Cenozoic,
and the Florissant locality was -2° north of its present loca-
tion. The lake sediments represent an estimated 5000 years
of deposition (McLeroy and Anderson, 1966). MacGinitie
(1953) recognizes 114 species distributed among the princi-
 Figure 6.7. Fossil plant locality
at Florissant, Colorado. People from
left to right: Richard A. Scott,
L. Imogene Doher, Alan Graham,
and Estella B. Leopold. Photograph
by Chester A. Arnold.

Table 6.6. Megafossils from latest Eocene-Early Oligocene Florissant flora of Colorado.

Bryophyta
Musci (mosses)
Grimmiaceae
Plagiopodopsis scudderi
P. cockerelliae

Pteridophytes
Equisetaceae Polypodiaceae
Equisetum florissantense Dryopteris guyottii

Gymnosperms
Gnetaceae Pinaceae cont.
Ephedra miocenica P. hambachi
Cupressaceae P. wheeleri
Chamaecyparis linguaefolia Taxaceae
Pinaceae Torreya geometrorum
Abies longirostris Taxodiaceae
Picea lahontense Sequoia affinis
P. magna
Pinus florissanti

Monocotyledon Angiosperms
Cyperaceae(?) Liliaceae
Cyperacites lacustris Smilax labidurommae
Gramineae Typhaceae
Stipa florissanti Typha lesquereuxi

continued
206 Late Cretaceous and Cenozoic History of North American Vegetation

Table 6.6. continued

Dicotyledon Angiosperms
Aceraceae Juglandaceae Rutaceae
Acer coloradense Carya libbeyi Ptelea cassiodes (see legend)
A. florissanti Lauraceae Salicaceae
A. heterodentatum Lindera coloradica Populus crassa
A. oregonianum Persea florissantia Salix coloradica
Anacardiaceae Sassafras hesperia S. libbeyi
Astronium truncation Leguminosae S. ramaleyi
Cotinus fraterna Caesalpinites acuminatus S. taxifolioides
Rhus lesquereuxi C. coloradicus Sapindaceae
R. obscura Cercis parvifolia Athyana haydenii
R. stellariaefolia Conzattia coriacea Cardiospermum terminalis
Schmaltzia vexans Leguminosites lespedezoides Dodonaea umbrina
Araliaceae Phaca wilmattae Koelreuteria alleni
Araliaceous fruits Phaseolites dedal Sapindus coloradensis
Oreopanax dissecta Prosopis linearifolia Thouinia straciata
Aristolochiaceae Robinia lesquereuxi Saxifragaceae
Aristolochia mortua Vicia Hydrangea fraxinifolia
Berberidaceae Meliaceae Philadelphus minutus
Mahonia marginata Cedrela lancifolia Ribes errans
M. obliqua Trichilia florissanti Simarubaceae
M. subdenticulata Moraceae Ailanthus americana
Betulaceae Morus symmetrica Staphyleaceae
Paracarpinus fraterna Myrtaceae Staphylea acuminata
Fagopsis longifolia Eugenia arenaceaeformis Styracaceae
Burseraceae Oleaceae Halesia reticulata
Bursera serrulata Osmanthus praemissa Thymelaeaceae
Caprifoliaceae Platanaceae Daphne septentrionalis
Sambucus newtoni Platanus florissanti Tiliaceae
Celastraceae Proteaceae Tilia populifolia
Celastrus typica Lomatia lineata Ulmaceae
Convolvulaceae Rhamnaceae Celtis mccoshii
Convolvulites orichitus Colubrina spireaefolia Ulmus tenuinervis
Euphorbiaceae Zizyph us florissanti Zelkova drymeja
Euphorbia minuta Zizyphus florissanti Verbenaceae
Fagaceae Rosaceae Holmskioldia speirii
Castanea dolichophylla Amelanchier scudderi (=Florissantia, Malvales)
Quercus dumosoides Cercocarpus myricaefolius Petrea perplexans (-
Q. knowltoniana Crataegus copeana Asterocarpinus perplexans,
Q. lyratiformis C. hendersoni Betulaceae)
Q. mohavensis C. nupta Vitaceae
Q. orbata Malus florissantensis Parthenocissus osbornii
Q. peritula M. pseudocredneria Vitis florissantella
Q. predayana Prunus gracilis
Q. scottii Rosa hilliae
Q. scudderi Vauquelinia coloradensis
V. liniara
Adapted from MacGinitie (1953). Ptelea cassiodes (Rutaceae) = Diplodipelta reniptera (Manchester and Donoghue, 1995). For Aceraceae, see revision
by Wolfe and Tanai (1987).

pal families Leguminosae (9 genera/10 species), Rosaceae
(7/11), Sapindaceae (6/6), Anacardiaceae (4/6), Pinaceae
(3/6), Ulmaceae (3/3), Rhamnaceae (3/3), Lauraceae (3/3),
and Saxifragaceae (3/3). The most abundant specimens are
identified as Fagopsis longifolia (Fig. 6.8; Betulaceae, now
Fagaceae; Manchester and Crane, 1983; 30%), Zelkova
drymeja (9.6%; =Cedrelospermum Lineatum; Manchester,
1989), Chamaecyparis linguaefolia (6.3%), Typha les-
quereuxi (5.8%), Populus crassa (5%), Rhus stellariaefolia
(2.8%), Staphylea acuminata (2.7%), Athyana haydenii
(2.3%), Sequoia afftnis (Fig. 6.9A; 2.2%), Cercocarpus myri-
caefolius (2.1%), Quercus scottii (1.5%), Rhus obscura
(1.3%), Pinus florissanti (1.3%), Sapindus coloradensis
(1.1%), Carya libbeyi (1.0%), and Bursera serrulata (1.0%).
Others include Ailanthus (Fig. 6.9C), Castanea (Fig. 6.9D),
Eucommia (Call and Dilcher, 1997), Holmskioldia speirii
(Verbenaceae; now Florissantia of the Bombacaceae-
Tiliaceae-Sterculiaceae complex; Manchester, 1992), Salix
(Fig. 6.9E), and the moss Plagiopodopsis (Fig. 6.9B). The
megafossil component of the Florissant flora has been up-

Figure 6.8. Fagopsis longifolia, Florissant flora. Reprinted
from MacGinitie (1953) with the permission of the Carnegie
Institution of Washington.
dated by Manchester (in press). The vegetation is inter-
preted as a rich mesic forest along stream and lake sides
and a drier evergreen oak-pine woodland on higher
ground, similar to the present dry savannalike community
of western Texas to northeastern Mexico (oak-laurel forest
of Asia or California woodland in western North America).
One of the characteristic features of the Florissant flora is
the small size of many leaves compared to their most simi-
lar modern species. The leaves are also mostly coriaceous
and not entire margined. Annual rainfall was estimated at
508-635 mm (20-25 in.) with ample summer rain, a pro-
nounced winter dry season, and a MAT of 18°C (65°F; i.e.,
mostly frost free). CLAMP yields a MAT of ~12.5°C (Wolfe,
1992b), and analysis of a new collection of 29 species gives
10.7 ± 1.5°C (K. M. Gregory, personal communication,
1995). The estimated precipitation trend between the
Kisinger Lakes (-48 Ma), Green River (45 Ma), and Floris-
sant (35 Ma) floras of Middle to Late Eocene age is shown in
Fig. 6.10. An unresolved problem is the paleoelevation of
the Florissant and other floras in western North America.
PALEOELEVATION ANALYSIS
Estimates of paleoelevation can be made on the basis of the
modern analog method. However, new ways are being de-
vised in an effort to more accurately determine and stan-
Middle Eocene through Early Miocene History 207
dardize the estimates (Axelrod, 1966a; Axelrod and Bailey,
1976; Gregory, 1996; Gregory and Chase, 1992; Gregory and
Mclntosh, 1996; Meyer, 1992; Wolfe, 1992b). Axelrod and
Bailey (1976; Axelrod, 1966a) developed the following
equation for estimating the paleoelevation of Late Eocene
to Middle Oligocene floras of the Rio Grande Depression:
(sea-level MAT - fossil flora MAT)
elevation = — + sea level,
terrestrial lapse rate
where sea-level MAT is the MAT of a sea-level fossil flora
coeval with the inland fossil flora (°C), fossil flora MAT is
the MAT of the inland fossil flora (°C), terrestrial lapse rate
is the vertical cooling rate of the atmosphere above a land
mass (°C/km), and sea level is the sea level at the time of
deposition of the inland fossil flora relative to modern sea
level (km). In this equation there is little accommodation
for climatic change or compensation for latitude, MAT is
based on modern analogs, and a lapse rate of 5.5°C/km is
assumed valid for almost all situations. Their results sug-
gest that paleoelevation in the Rio Grande Depression re-
gion increased by ~2000 m in the 6-8 Ma interval between
the Eocene (38-40 Ma) and the Early Oliogcene (32 Ma).
Lapse rate is the change in temperature as a function of
latitude or altitude. The altitudinal lapse rate has been cal-
culated for various sites throughout the world and aver-
 208 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 6.9. (A) Sequoia affinis, Florissant flora. (B) Plagiopodopsis cockerelliae, Florissant flora (C) Ailanthus
amencana, Florissant flora. (D) Castanea dolichophylla, Florissant flora. (E) Salix coloradica, Florissant flora
Reprinted from MacGinitie (1953) with the permission of the Carnegie Institution of Washington.

aged into a worldwide mean of 5.0-6.0°C /km. However,
the figure is higher for windward than leeward slopes and
varies within a site over time, depending on the strength of
incursions of frigid Arctic air that in turn are influenced by
rising mountains and plateaus. Other factors include high
base levels, fog, cold-air drainage, and local upwelling.
Meyer (1986) determined temperature differences be-
tween stations in the same vicinity as the fossil flora and
applied a lapse rate of 5.9°C /km (more widely >4°C and
<8°C ). By this calculation Eocene elevations in the Rio
Grande Depression region were already 3000-4700 m.
Wolfe derived lapse rates
[B]y calculating temperature differences between coastal
and upland and (or) interior areas, which is the basis for
the methodology proposed by Axelrod (1966[a]). I have,
moreover, included all lower altitude, interior stations,
most of which were excluded by Meyer (1986); such
sites probably are present in topographic analogues of
Tertiary depositional basins, which are basic to paleo-
altitudinal estimates. (1992b)
A lapse rate of 2.0-3.0°C is suggested for western conter-
minous North America (Wolfe, 1992b). The CLAMP pro-
gram can be applied to leaf physiognomic data to refine es-
timates of MAT for coastal and upland fossil floras. It
should be noted, however, that the MATs for coastal fossil
floras are estimates (CLAMP= ± 2°C), the MATs for upland
floras are estimates, and the lapse rate is an approximation.
Thus, a range of values is presently possible for the calcu-
lated paleoelevations.
The paleoelevation of the Florissant flora was estimated
at -900 m based on the modern analog method (MacGini-
tie, 1953). Using the newer method, and basing estimates

I I
of MAT on foliar physiognomy, Wolfe (1992b) compares
the MAT of the inland flora with that of sea-level flora(s) of
the same age and applies a lapse rate of 3.0°C/km, consid-
ered most accurate when beginning with a high base level
(Wolfe, 1994b). Meyer (1992) estimates paleoelevation by
projecting the sea-level MAT inland to the MAT of the in-
land flora (e.g., Florissant) using the worldwide average
lapse rate of ~5.9°C /km and incorporating other compen-
sating factors. These various approaches give an estimated
paleoelevation for the Florissant flora of 2300-3300 m
(Gregory, 1994), 2450 m (Meyer, 1986, 1992), and 2700-
2900 m (Wolfe, 1992b), or essentially modern elevations by
the Eocene-Oligocene. (The present elevation is ~2500 m.)
In another study, Gregory and Mclntosh (1996) used the
Goshen flora (MAT = 19.5°C) as the sea-level assemblage
in calculating the paleoelevation of the Pitch-Pinnacle
flora. This flora is from the west flank of the Sawatch
Range, Colorado, west of Florissant, and is dated at
32.9-29 Ma. Using foliar physiognomy as a basis, Early
Oligocene paleoelevation is estimated at near-modern val-
ues. In the Basin and Range province paleoelevations were
calculated for two Late Eocene floras in northeastern
Nevada (Povey et al., 1994). Sea-level data was based on
Middle Eocene through Early Miocene History 209
Eocene Rocky Mountains floras.
tion estimated from Middle and Late
eocene rocky mountaions
Reprinted from Leopold et al. (1992)
with the permission of Princeton Uni-
versitreeepress
fossil floras in the Puget Group of Washington and MAT
was calculated using CLAMP. The results also suggest pa-
leoelevations similar to those of today.
Estimates of paleoelevation involving MATs and lapse
rates (2-5.9°C/km) are presently imprecise. In addition to
the approximations used in the calculations, another prob-
lem is that changes in either elevation or climate can pro-
duce similar signals from fossil floras; in much of Cenozoic
North America both were occurring simultaneously. A pos-
sible way of avoiding errors introduced by uncertainties in
MAT and lapse rates is to use the paleobotanical data to
infer enthalpy (a measure of the energy content of a system
per unit mass). This approach yields a paleoelevation for
the Florissant flora of 2.9 ± 0.7 km (Forest et al., 1995), and
paleoelevations in the Rio Grande Rift area of New Mexico
are estimated at -3000-4700 m. Gregory-Wodzicki (1997)
estimates the paleoelevation of the Late Eocene House
Ranch flora (31.4 ± 0.5 Ma) of western Utah at 3.6 ± 0.7 km
(MAT-based) or 2.9 ± 1.5 km (enthalpy-based; presently 1.7
km). Analyses of 12 Middle Miocene floras from western
Nevada indicate a paleoelevation of ~3 kms above sea level
at 15-16 Ma. This is 1-1.5 km higher than at present, and
the difference is attributed to collapse by -13 Ma (Wolfe et
210 Late Cretaceous and Cenozoic History of North American Vegetation

al., 1997). By all these data, sections of the Rocky Moun- The Metzel Ranch flora in the Upper Ruby River Basin
tains in central Colorado, southwestern Montana, New of southwestern Montana, just to the east of the Beaver-
Mexico, and in the Basin and Range Province locally had head Basins, was also considered Oligo-Miocene in age
reached at least modern elevations by the latest Eocene and (Becker, 1972); but recent estimates place it, along with the
would have been adequate to cast rainshadows over parts Mormon Creek (Becker, 1960) and York Ranch (Becker,
of the eastern slopes and High Plains and to influence at- 1973) floras of Montana, in the Early Oligocene Orellan
mospheric circulation. These data imply a reduced role for NALMA (32.2-30.8 Ma, Wing, 1987; Fig. 3.13). It is dis-
uplift in Late Cenozoic cooling (Chapter 2, Orogeny). tinct in composition and paleoecological import from the
Beaverhead Basins and Ruby floras. It differs primarily in
THE FLORA OF THE BEAVERHEAD BASINS in southwestern Montana the poorer representation of gymnosperms and in the ab-
was assigned to the Oligo-Miocene (Becker, 1969), but it sence of oaks. The prominence of Rosaceae (10.9%), Legu-
is now placed in the latest Eocene to earliest Oligocene and minosae (6.53%), and Rhamnaceae (6.53%), along with the
is approximately coeval with the Florissant flora (Wing, presence of Juniperus, Gramineae, Crataegus, and Maho-
1987; S. L. Wing, personal communication, 1994). The nia, suggest dryness and the absence of Abies and Picea in-
change is based on Chadronian or older vertebrate faunas dicates low elevations in the immediate vicinity. The Met-
from the Renova Formation that includes the Beaverhead zel Ranch flora is an example of the local drier conditions
Basins flora. It comprises 110 genera and 160 species as de- and the move toward better definition and expansion of a
scribed by Becker (1969). Prominent genera include Quer- shrubland/chaparral-woodland-savanna to the west of
cus (nine species); Mahonia (eight); Acer (six); Salix (four); the Rocky Mountains as the Pacific sierras just began to af-
and three species each of Abies, Picea, Betula, Cassia, Cer- fect moisture regimes inland. The trend was intensified as
cocarpus, Fraxinus, and Zelkova. Taxonomic revisions for more pronounced highs settled over the region.
Ulmus and Zelkova in Middle to Late Tertiary floras of Another Late Eocene to Early Oligocene flora in south-
western North America have been presented by Tanai and western Montana is preserved in intermontane lacustrine
Wolfe (1977). The communities include aquatic and lake- shales of the Ruby Basin. The Ruby flora is slightly younger
shore vegetation (Taxodium, Typha), flood-plain and (Whitneyan NALMA; 30.8-29.2 Ma; Wing, 1987) and
humid forest (Chamaecyparis, Sequoia, Thuja, Cercidi- somewhat drier than the Mormon Creek, Metzel Ranch,
phyllum, Salix), deciduous and mixed deciduous hard- and York Ranch floras. Becker (1961) recognized 61 genera
wood-coniferous forest on the slopes (Acer, Corylus, and 82 species. Forty percent of the species also occur in
Ulmus}, a high montane forest (Abies, Picea), and a subhu- the Florissant flora. The two floras reflect a similar vegeta-
mid chaparral formation on drier exposed slopes (Juni- tion, but the Ruby flora is slightly more mesic and upland
perus, Arctostaphylos, Berberis, Cercocarpus, Mahonia, in aspect. Genera present in the Ruby vegetation but absent
Potentilla). Plants presently of eastern North America, from the Florissant include Glyptostrobus, Metasequoia,
western North America, and Asia (Ginkgo, Glyptostrobus, Pseudotsuga, Alnus, Betula, Cornus, Fagus, Fraxinus, My-
Metasequoia, Ailanthus, Eucommia) are also represented. rica, and Nyssa. Tropical warm-temperate and drier ele-
The paleoelevation of the lake basin is estimated at ments present in the Florissant and absent from the Ruby
500-700 m (1500-2000 ft), and highlands were up to 2000 assemblage include Ephedra, Bursera, Cedrela, Lindera,
m (6000 ft) in the vicinity. The presence of Sequoia and Ce- Persea, Prosopis, and Trichilia. All of these Late Eocene to
drela suggests minimum temperatures were at to just above Early Oligocene floras from southwestern Montana preserve
freezing, and annual precipitation is placed at 1000-1300 a mixed coniferous and broad-leaved deciduous forest and
mm (40-50 in.). shrubland that grew under temperate to dry-temperate con-
One specimen from the Beaverhead Basins flora, Vigu- ditions and show no clear climatic trend during the inter-
iera cronquistii (Compositae-Asteraceae), warrants special val (Wing, 1987; Wolfe, 1992a). The CLAMP estimate of
comment. This is an important identification because it MAT is 11-12°C with an annual range of 16°C.
would be the oldest megafossil record of the family in The Late Eocene (-40 Ma) Copper Basin flora in north-
North America (the oldest verified pollen is Oligo- eastern Nevada reveals that the higher elevations and more
Miocene; Graham, 1996), and it has been used as evidence moist conditions west of the Rockies reached by the Mid-
for positioning the Heliantheae as the primitive tribe in the dle Eocene (Wind River, Green River vs. Thunder Moun-
Compositae. Also, modern species of Viguiera are mostly tain, Germer floras) persisted through the Late Eocene and
polyploids, a relatively advanced cytological feature, later time. The Copper Basin flora (Axelrod, 1966b), in con-
which would favor arguments that the family had a much trast to the Florissant, is a microthermal mixed deciduous
earlier origin. A reexamination of the specimen revealed hardwood-coniferous assemblage. The 42 species repre-
that although it is composite-like in aspect (only bracts of sent three communities adjacent to a small lake. The bor-
the supposed capitulum are preserved), the fossil "cannot dering vegetation consisted of Acer, Alnus, Amelanchier,
be considered unequivocally to be the remains of a com- Crataegus, Lithocarpus, Mahonia, Prunus, and Salix. The
posite" (Crepet and Stuessy, 1978). Its affinities are pres- slope forest included Cephalotaxus, Chamaecyparis, Pseu-
ently unknown. dotsuga, Sequoia, Acer, Aesculus, Euonymus, Prunus, Rho-

dodendron, Sassafras, and Ulmus. The montane conifers
were Abies, Picea, and Pinus, associated with Acer and
Ribes, and a subalpine community of Picea, Pinus, Larix,
and Tsuga. Annual precipitation is estimated at 1270-
1524 mm (50-60 in.), MAT at 11°C (51°F; cool temperate),
and the elevation -1200 m (3600 ft; modern elevation 7100
ft). The study is of interest because at the time, mixed de-
ciduous hardwood-coniferous forests were unreported
from the Eocene of the western United States. The Copper
Basin and Republic floras (see later discussion) show that
the community was developing by the Middle Eocene
(viz., Abies, Picea}. The Bull Run flora (Axelrod, 1966b,c)
lies just to the west of Copper Basin. It is probably some-
what younger in age, but it also preserves a near pure west-
ern montane coniferous forest of Abies, Picea, Pinus,
Pseudotsuga, Tsuga, Chamaecyparis, and Thuja at the
highest elevations (1250-1500 m) and a few moist decidu-
ous angiosperms with small leaf indices further down-
slope.
Until recently, the Middle Eocene vegetation throughout
southern California had to be generalized by extrapolation
from floras well outside the area. Tropical rain forest was
present along the coast to the north (Wolfe, 1985, 1989),
and it could have extended southward. However, modeling
results by Parrish et al. (1982) simulate precipitation in the
Lutetian (50-42.1 Ma) as somewhat lower in southern Cal-
ifornia and northern Mexico than in areas to the north.
This is consistent with evidence from sedimentology, ver-
tebrate paleontology, and a fossil pollen record from the
San Diego area (Elsinore flora) that shows some winter dry-
ing, and possibly a cooling trend, beginning in the Middle
and Late Eocene (Frederiksen, 1991). Black fusinite and
dark brown semifusinite from the uplands are present in
the samples, indicating forest fires (Cope, 1981). The cli-
matic deterioration between the Early Lutetian and the Pri-
abonian shows a similar stepwise pattern as recorded in
the southeastern United States. The limited evidence sug-
gests a winter-dry mixed deciduous hardwood-coniferous
forest in the uplands and the beginning of a complex trend
leading to the replacement of the Paleocene to earliest
Lutetian tropical to paratropical lowland vegetation by
more seasonally dry communities. The dry tropical forest
proposed by Axelrod (1979) for the early Lutetian, based
on some questionable pollen identifications (Ting, 1975;
W. S. Ting, unpublished data; see also Frederiksen, 1991),
probably had not yet developed. Analysis of paleosols in
San Diego County and northwestern Mexico reveal an an-
nual rainfall of 1250-1900 mm (Abbott, 1981; Peterson
and Abbott, 1979), and a megafossil assemblage from the
Lower Lutetian Torrey Sandstone of San Diego is inter-
preted as growing under a paratropical climate with an an-
nual rainfall of 1200-1500 mm (Myers, 1990).
This drying trend is evident in the early Lutetian by a
turnover peak (first pollen event) wherein the last appear-
ances of Milfordia (Restionaceae), Diporoconia (Palmae?),
Lanagiopollis (Alangium), and Triatriopollenites (probably
Middle Eocene through Early Miocene History 211
Myrica) correspond to the first records of Ruellia (?; Acan-
thaceae), Bursera (Burseraceae), Pachysandra-Sarcococca
(Buxaceae), Lonicera type (Caprifoliaceae), Onagraceae,
other Restionaceae, Chiranthodendron- Fremontodendron-
Triumfetta (Bombacaceae-Sterculiaceae-Tiliaceae), Cyril-
laceae(?), and Juglandaceae. The drying may have intensi-
fied during the Middle and Late Lutetian as shown by
caliche and salt crystal formation, which suggest an esti-
mated annual rainfall of ~500-750 mm in the lowlands
(500-1000 mm based on faunal evidence; Axelrod, 1979;
Novacek and Lillegraven, 1979). The vegetation during the
Middle Lutetian has been interpreted as a form of savanna,
but with understory shrubs and herbs rather than grasses
and more mesic trees growing as gallery forests. The plant
microfossil evidence confirms the unusual nature of the
"savanna" here, as in the Green River Formation, in that
grasses were not prominent in the understory as in modern
savannas. Pollen evidence further reveals a greater diver-
sity of vegetation with more trees [gallery forests; Palmae,
Bombacaceae—Sterculiaceae —Tiliaceae, Eucommiaceae,
Fagaceae(?), Juglandaceae, Myrtaceae(?), Symplocaceae(?),
Ulmaceae] than suggested by the sedimentological and fau-
nal evidence from the interplains, which emphasizes
greater aridity. Coarse detritus from the highlands indicates
drier conditions there with a longer and more pronounced
dry season.
By the Bartonian the drying trend, and perhaps cooling
temperatures, was having a more pronounced effect on the
vegetation of southern California as evidenced by the MEDD
(second pollen event), which is also present in the south-
eastern United States. The last appearances during this in-
terval include Acanthaceae, Aquifoliaceae, Bombacaceae,
Euphorbiaceae, Palmae(?), Symplocaceae(?), Tiliaceae, and
Ulmaceae(?). In contrast to the southeast, however, the
MEDD was not followed by abundant introductions (e.g.,
there is no evidence of grasses in the Eocene of southern
California), so the net decrease in pollen diversity is greater.
In the Lower Priabonian a third pollen event is ex-
pressed by a rapid increase in Quercoideae forms, as in the
southeastern United States, which is paralleled by the be-
ginning of redbed deposition in the Sespe Formation of
Ventura County, California (Dickinson and Leventhal, 1988).
The fourth event is a continued decline in diversity and
corresponds to the Late Eocene-Early Oligocene cooling.
The culmination of these trends produced a winter dry
paratropical forest or a type of savanna with gallery forests
along streams and in the lowlands and a more distinctly
seasonally dry mixed coniferous forest in the uplands. This
marks an early stage in the development of the Madrean
(scrubland/chaparral-woodland-savanna) vegetation of
the present arid southwestern United States and north-
western Mexico. The rainfall pattern was summer wet, so
that a major change after the Eocene was the development
of the summer dry Mediterranean climate that has charac-
terized the region since Plio-Pleistocene times.
The late Middle Eocene Susanville flora of northern Cal-
212 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 6.11. Floral stage names proposed for western North America (Wolfe, 1981a) and
some radiometric dates cited for the floras. See text for revised radiometric dates.

ifornia (40° N latitude) includes notophyllous to mega-
phyllous broad-leaved evergreens (some with finely serrate
margins) that are interpreted as a coastal tropical rain for-
est growing under a MAT of ~27°C (Wolfe, 1978,1985). The
various Platanaceae remains (Macginitiea, Macginicarpa,
Platananthus) described from the Late Paleocene Joffre
Bridge flora also occur in several floras from California
(Chalk Bluffs, late Early Eocene), Oregon (Clarno Forma-
tion, West Branch Creek locality, Middle to Late Eocene),
and Washington (Steels Crossing, Middle Eocene; Man-
chester, 1986).
The latest Eocene-Early Oligocene La Porte flora (Pot-
bury, 1935; 32-33.2 Ma), south of Susanville in northwest-
ern California, is similar in aspect, but it has not been re-
vised since the original publication. The MAT was
originally estimated at ~24°C, and the CLAMP MAT esti-
mate is ~22.3°C (Wolfe, 1993). These values are near those
of the Comstock and Goshen floras 8° latitude to the north,
suggesting deposition at a slightly higher elevation (Wolfe,
1992b). The Lower Cedarville flora of far northeastern Cal-
ifornia is older but also Late Eocene in age. Wolfe (1981a)
has proposed a series of botanically defined floral stages
based on these Pacific Northwest communities (but see Ax-
elrod, 1987). The stages for the period -50-16.3 Ma are
shown in Fig. 6.11, along with some floras discussed in the
text and their original radiometric dates (see text for re-
vised dates).
The Middle-Late Eocene Clarno flora is located in the
John Day Basin of north-central Oregon. One locality
dated at -44 Ma consists of numerous fruits and seeds
with some leaf and wood material and is known as the
"nut beds" by local collectors. These stream and lake delta
deposits have been studied most extensively among the
Clarno localities (Manchester, 1990,1994; Scott, 1954) and
comprise 145 genera and 173 species (Table 6.7). Most are
trees and lianas. Among the 69 species for which good es-
timates of growth form could be made, 43% are lianas
from such families as the Icacinaceae, Menispermaceae,
and Vitaceae. About 52% of the dicot leaves are entire
margined, and broad-leaved evergreens are abundant. The
flora is tropical to paratropical in composition with only a
few temperate genera (e.g., Ginkgo, Pinus, Taxus, Em-
menopterys, Parthenocissus). Twenty-four percent of the
genera are found in Early to Middle Eocene floras of Eu-
rope, consistent with geological and other biological evi-
dence for land connections across the North Atlantic and
Table 6.7. Composition of Middle Eocene Nut Beds flora, Clarno Formation, Oregon.

Gymnosperms
Pinaceae Taxaceae
Pinus sp. Diploporus torreyoides
Taxus masonii
Torreya clarnensis

Monocotyledon Angiosperms
Musaceae Palmae
Ensete oregonense Sabal bracknellensis
S. jenkinsii

Dicotyledon Angiosperms
Actinidiaceae Juglandaceae Sabiaceae
Actinidia oregonensis Juglandeae Meliosma beusekomii
Alangiaceae Cruciptera simsonii M. bonesii
Alangium eydei Juglans clarnensis M. elongicarpa
A. rotundicarpum Engelhardieae M. cf. jenkinsii
Anacardiaceae cf. Palaeocarya clarnensis M. leptocarpa
Pentoperculum minimus Platycaryeae Sabia americana
Rhus rooseae Paleoplatycaryal hickeyi Sapindaceae
Annonaceae Lauraceae Deviacer wolfei
Anonaspermum cf. pulchrum Laurocalyx wheelerae Palaeoallophylus globosa
A. bonesii Laurocarpum hancockii P. gordonii
A. rotundum L. nutbedensis Sapotaceae
Araliaceae L. raisinoides Bumelial globosa
Paleopanax oregonensis Lindera clarensis B.I subangularis
Betulaceae Leguminosae Schisandraceae
Coryloides hancockii Leguminocarpon sp. Schisandra oregonensis
Kardiasperma parvum Lythraceae Staphyleaceae
Burseraceae Decodon sp. Tapiscia occidentalis
Bursericarpum oregonense Magnoliaceae Symplocaceae
B. sp. Magnolia muldoonae Symplocos nooteboomii
Cornaceae M. paroblonga Theaceae
Cornus clarnensis M. tiffneyi Cleyera grotei
Langtonia bisculcata Menispermaceae Ulmaceae
Mastixia sp. Coscineae Celtidoideae
Mastixioidiocarp um Oregon ense Anamirta leiocarpa Amphananthe maii
Mastixicarpum occidentale Menispermeae Celtis burnhamae
Nyssa scottii Diploclisia auriformis C. sp.
N. spatulata Eohypserpa scottii Trema nucilecta
N. sp. Davisicarpum limacioides Ulmoideae
Fagaceae Palaeosinomenium venablesii Cedrelospermum lineatum
Castanopsis crepetii Tinosporeae Vitaceae
Quercus paleocarpa Atriaecarpum clarnense Ampelocissus auriforma
Flacourtiaceae Calycocarpum crassicrustae A. scotti
Saxifragispermum tetragonalis Chandlera lacunosa Ampelopsis rooseae
Hamamelidaceae Curvitinospora formanii Parthenocissus angustisulcata
Fortunearites endressii Odontocaryoidea nodulosa P. clarnensis
Hydrangeaceae Thanikaimonia geniculata Vitis magnisperma
Hydrangea knowltonii Tinomiscoidea occidentalis V. tiffneyi
Icacinaceae Tinospora elongata
lodeae T. hardmanae
Comicilabium atkinsii Platanaceae
lodes chandlerae Macginicarpa glabra
I. multireticulata Platanus hirticarpa
lodicarpa ampla Tanyoplatanus cranei
I. lenticularis Rosaceae
Phytocreneae Prunus weinsteinii
Palaeophytocrene hancockii P. olsonii
P. pseudopersica Rubiaceae
Pyrenacantha occidentalis Emmenopterys dilcheri

Adapted from Manchester (1994); Incertae Sedae are not included. For Mastixia—Mastixioidiocarpum—Mastixicarpum (Cornaceae), see Tiffney and
Haggard (1996).
214 Late Cretaceous and Cenozoic History of North American Vegetation
the North Pacific during the Early Tertiary. The flora is pri-
marily Asian in affinity, however, and includes Actinidia,
Alangium, Ampelocissus, Celtis, Cornaceae (Cornus,
Langtonia, Mastixia, Nyssd), Decodon, Ensete (Manchester
and Kress, 1993; the only herbaceous angiosperm), Eu-
commia Florissantia (malvalean affinity; Manchester,
1992), Hydrangea, lodes, Juglandaceae (Cruciptera; Man-
chester, 1991), Magnolia, Meliosma, Musa, Platanaceae,
Quercus, Sabia, Schisandra, and Tiliaceae (Pteleae-
carpum; Biizek et al., 1989; Kvacek et al., 1991). The latter
fossil occurs at only one probable Clarno locality and is
more characteristic of the Bridge Creek flora. It is similar to
the extant Chinese genus Craigia of the Tiliaceae. The pres-
ence of Quercus is noteworthy because this record, to-
gether with that from the Middle Eocene Chalks Bluff flora
(MacGinitie, 1941; Wolfe, 1973), are among the earliest oc-
currences of the genus. The southeast Asian genus Eneste
(Musaceae) presently grows where the MAT is between 20
and 28°C and the annual range is between 2 and 19°C (viz.,
frost free). The CLAMP MAT estimate is 16°C. Among the
vertebrate fossils is one crocodilian. Because of the prox-
imity of the flora to the paleocoast of Oregon ~100 km to
the west and the absence of the Cascade Mountains and
Coast Ranges, moisture was greater and more uniformly
distributed than it was to the east of the Rocky Mountains
where the vegetation was tropical but seasonally dry
(Leopold and MacGinitie, 1972; Wing, 1987; each side of
the hatched area in Fig. 5.8). Some seasonality in rainfall
may be inferred from the woods (e.g., Vitaceoxylon tiffneyi,
V. carlquistii; Vitaceae) that contain growth rings (Wheeler
and LaPasha, 1994). However, in Clarnoxylon blanchardii
(Juglandaceae; Manchester and Wheeler, 1993), although
growth rings are present, the vessel elements lack helical
thickenings and are thus more characteristic of tropical
than temperate species (Van der Graaff and Baas, 1974).
The Late Eocene Goshen (-31 [39] Ma; Chaney and San-
born, 1933) and Comstock (Sanborn, 1935) floras of west-
central Oregon have not been revised since the original
publications, but they generally represent moist forests
similar to the modern upland vegetation of southeastern
Mexico (Allophylus, Annona, Astronium, Cupania, Drimys,
Ficus, Hydrangea, Ilex, Inga, Nectandra, Ocotea, Platanus,
Quercus, Sapium, Siparuna, Smilax). The specimen de-
scribed as Viburnum palmatum (Chaney and Sanborn,
1933) is now Florissantia ashwillii (Manchester, 1992). The
CLAMP MAT estimate is 20°C (Wolfe, 1992a).
The early Middle Eocene Republic flora of northeastern
Washington (Klondike Mountain Formation) was deposited
at a paleoelevation of 700-900 m and is dated at 48-49 Ma
(Wolfe and Wehr, 1987; see also Wehr, 1995; and a series of
papers in Washington Geology). The depositional basin at
Republic was associated with a raised structure called the
Eocene Interior Arc (Myers, 1996). The flora further em-
phasizes differences between the landscape and vegetation
of the eastern Rocky Mountains-Great Plains area and re-
gions to the west (Fig. 5.8). In the east, a microthermal ele-
ment was present but not dominant in the primarily tropi-
cal to subtropical seasonally dry climate and generally low
elevations. To the west, microthermal elements are present
and diverse due to the higher paleoelevation (Axelrod,
1966a). The vegetation was a microthermal mixed decidu-
ous hardwood-coniferous forest (Table 6.8). The MAT is
estimated at 12-13°C, and the mean annual range of tem-
perature (MART) at 5-6°C.
The Republic flora is the oldest known assemblage with
co-occurring members of a western montane coniferous
forest of Pinaceae (Abies, Picea, Pinus, Pseudolarix, Tsuga)
and Cupressaceae (Chamaecyparis, Thuja). The oldest
known Betula related to B. papyrifera—B. occidentalis and
the oldest Tilia are from the Middle Eocene Republic flora.
Also present is "Holmskioldia" speirii (Verbenaceae), now
considered as Florissantia quilchensia (Manchester, 1992).
According to Wing (1987), the early diversification of many
modern microthermal lineages likely took place during
the Eocene in the volcanic highlands of the northwestern
United States (see also Wolfe, 1986, 1987). The western
Montane Coniferous Forest association further developed
and continued to expand in Late Eocene and Neogene
times. Although the number of floras is few, it is probable
that the northern Rocky Mountains were occupied primar-
ily by a coniferous forest for most of the Oligocene and
later times, replacing the older megathermal evergreen veg-
etation. The Middle Eocene Republic flora records the be-
ginnings of the transition.
An extensive flora of Middle Eocene age (-45-50 Ma) is
preserved in the Allenby Formation near Princeton, south-
ern British Columbia, Canada (-49° N). Results of the study
are being published separately for each taxon by Ruth
Stockey and coworkers and when completed a detailed pic-
ture will be available of the Middle Eocene vegetation of
southwestern Canada (Pigg and Stockey, 1996). The mem-
bers identified to date are listed in Table 6.9. Provisional
identifications pending further study include Abies, Acer,
Aesculus, Castanea, Cercidiphyllum, Fagus, and Ulmaceae
(Crane and Stockey, 1987). [In addition to references listed
in Table 6.9, see also Cevallos-Ferriz et al. (1991), Erwin
(1987), and Erwin and Stockey (1991b).] Other fossils are
of unknown family affinity (e.g., Princetonia allenbyensis,
Magnoliopsida, Fig. 6.12; Ethela sargantiana, monocotyle-
don, Erwin and Stockey, 1992; Eorhiza arnoldii). The flora
is presently interpreted as a shallow, shoreline, lacustrine
ecosystem (marsh, swamp, lake margin) growing under
moist, subtropical to warm-temperate conditions and it
seems to represent a mixed deciduous hardwood (with de-
ciduous coniferous elements)-evergreen coniferous forest
that extended southward through Washington (Republic
flora) and northeastern Nevada (Copper Basin flora). In the
Middle Eocene generally across the northern region other
plants include Glyptostrobus, Larix, Picea, Taxodium,
Alnus, Betula, Nyssa, Platanus, Populus, Pterocarya, and
Quercus (Axelrod, 1984).
In the late Middle Eocene, megathermal tropical and
Table 6.8. Composition of early Middle Eocene Republic flora, northeastern Washington.

Gymnosperms
Cephalotaxaceae Pinaceae Taxus type
Cephalotaxus type Abies milleri Torreya
Cupressaceae Picea Taxodiaceae
Chamaecyparis Pinus Cryptomeria type
Thuja Pseudolarix americana Glyptostrobus type
Ginkgoaceae Tsuga Metasequoia occidentalis
Ginkgo biloba Taxaceae Sequoia
Amentotaxus type

Monocotyledon Angiosperms
Musaceae
Ensete

Dicotyledon Angiosperms
Aceraceae Icacinaceae Sabiaceae
"Acer" arcticum (Deviacer, Palaeophytocren e Meliosma
Sapindaceae) Iteaceae Sabia
Acer sp. Itea sp. Salicaceae
Anacardiaceae Juglandaceae Populus
Rhus malloryi Carya? sp. Salix
Betulaceae Carya/Juglans Sapindaceae
Alnus parvifolia Cruciptera Bohlenia americana
Betula leopoldae Pterocarya sp. Deviacer ("Acer" arcticum)
Corylus Lauraceae Koelreuteria arnoldi
Paleocarpinus Phoebesp. Smilacaceae
aff. Corylus Sassafras hesperia Smilax
aff. Carpinus Leguminosae Sterculiaceae
Bignoniaceae Menispermaceae Florissantia quilchenensis
Burseraceae Calycocarpum Theaceae
Barghoornia oblongifolia Myricaceae Ternstroemites
Cercidiphyllaceae Comptonia columbiana Tiliaceae
Cercidiphyllum obtritum Myrtaceae Craigia
Cornaceae Paleomyrtinaea Tilia johnsoni
Cornus sp. Nymphaeaceae Trochodendroid group
Davidiaceae Nuphar Cercidiphyllum
Tsukada davidiifolia Olacaceae Joffrea
Eucommiaceae Schoepfia republicensis Nordenskioldia
Eucommia montana Platanaceae Trochodendron
Macginicarpa Ulmaceae
Castaneophyllum Macginitiea gracilis Cedrelosperm um
Fagopsis undulata Rosaceae Ulmus
Grossulariaceae Crataegus Verbenaceae?
Ribes Photinia pageae "Holmskioldia" speirii
Hamamelidaceae Potentilla pageae Vitaceae
Corylopsis Prunus Vitis
Langeria magnifica Spiraea Incertae sedis
Liquidambar Rubiaceae Calycites ardtunesis
Hydrangeaceae cf. Emmenopterys Republica hickeyi
Hydrangea Pteronepelys wehri

Adapted from Gandolfo (1996), Schorn and Wehr (1996), Wehr and Hopkins (1994), Wehr and Manchester (1996), and Wolfe and Wehr
(1987). Cruciptera (Juglandaceae) is added by Manchester (1991); Eucommia montana is added by Call and Dilcher (1997).
Table 6.9. Composition of Middle Eocene Princeton Chert flora (Allenby Formation) near Princeton, southern British
Columbia, Canada.
Taxon Reference

Pteridophytes
Blechnaceae
Blechnoid fern
Dryopteridaceae
Diplazium Rothwell et al. (1994)
Onocleoid fern
Osmundaceae
cf. Osmunda
Polypodiaceae
Dennstaedtiopsis aerenchemata Basinger and Rothwell (1977)

Gymnosperms
Pinaceae
Pinus andersonii Stockey (1984)
P. arnoldii Miller (1973); Stockey (1984)
P. princetonensis Stockey (1984)
P. similkameenensis Miller (1973), Stockey (1984)
P.sp. Phipps et al. (1995)
Taxodiaceae
Metasequoia milled Basinger (1981, 1984),
Basinger and Rothwell (1977),
Rothwell and Basinger (1979)

Monocotyledon Angiosperms
Alismataceae
Heleophyton helobiaeoides Erwin and Stockey (1989)
Araceae
Keratosperma allenbyensis Cevallos-Ferriz and Stockey (1988a)
Arecaceae (Palmae)
Uhlia allenbyensis Erwin and Stockey (1991a, 1994)
Juncaceae - Cyperaceae
Ethela sargantiana Erwin and Stockey (1992)
Liliaceae
Soleredera rhizomorpha Erwin and Stockey (1991a)

Dicotyledon Angiosperms
Cornaceae
Mastixicarp um Wehr (1995)
Grossulariaceae
Ribes Cevallos-Ferriz (1995)
Lauraceae Sun and Stockey (1991)
Lythraceae
Decodon allenbyensis Cevallos-Ferriz and Stockey (1988b)
cf. Lythrum Cevallos-Ferriz and Stockey (1988b)
Myrtaceae
Paleomyrtinaea princetonensis Pigg et al. (1993)
Magnoliaceae
Liriodendroxylon princetonensis Cevallos-Ferriz and Stockey (1990b)
Nymphaeaceae
Allenbya collinsonae Cevallos-Ferriz and Stockey (1989)
Rosaceae
Paleorosa similkameenensis Basinger, 1976, Cevallos-Ferriz et al. (1993)
Prunus allenbyensis Cevallos-Ferriz and Stockey (I990c)
Prunus (types 1,2,3) Cevallos-Ferriz and Stockey (1991)
Sapindaceae
Wehrwolfea striata Erwin and Stockey (1990)
Vitaceae
Ampelocissus similkameenensis Cevallos-Ferriz and Stockey (I990a)
Unknown Dicotyledonae
Eorhiza arnoldii Robison and Person (1973)
Stockey and Pigg (1994)
Princetonia allenbyensis Stockey (1987)
Stockey and Pigg (1991)
Adapted from Pigg and Stockey (1996) and Cevallos-Ferriz et al. (1991). Taxa without references are R. A. Stockey, personal communication, 1996).
 Middle Eocene through Early Miocene History 217

leaf-size index is 57. A summary of high-latitude floras and

Figure 6.12. Longitudinal sec-
tion through the fruit Prince-
tonia allenbyensis, Princeton
flora, British Columbia.
Reprinted from Stockey and
Pigg (1991) with the permis-
sion of Elsevier Science-NL.
paratropical forests were reduced in extent as temperatures
(especially minimum winter temperatures) continued to
decline. The Paleogene floras from the main part of Alaska
(Wolfe, 1977, 1992a) record a change from a late Early
Eocene Asian paratropical vegetation to a Middle Eocene
cooler broad-leaved evergreen subtropical community (En-
gelhardia type, Cedrela-Melia, Ilex; CLAMP MAT estimate
~17.2-16.2°C, originally reported as ~18°C) to an Oligo-
cene temperate forest (Metasequoia, Alnus; 4.5-7°C), for a
decline of 8-10.5°C . The Eocene coast was fringed with
evergreen Lauraceae, and the MAT is estimated at ~16°C .
The Late Eocene Rex Creek flora of Alaska includes Meta-
sequoia, Acer, Alnus, Betula, Carya, Castanea, Cedrela,
Fagus, Juglans, Liquidambar, Pachysandra, Platanus,
Prunus, Pterocarya, Quercus, and Ulmus. The CLAMP esti-
mate is 12.3°C. These values are broadly consistent with a
MAT of >13°C for the notophyllous broad-leaved evergreen
forest (oak-laurel type) from the Late Eocene Puget Group
flora of northeastern Washington (Burnham, 1994). There
entire leaf margin species average 50%, and the average
the inferred climates is given in Table 6.10.
The climatic patterns deduced from the proxy data of
plant fossils just described are also consistent with recent
independent context information from clay minerals
(Robert and Chamley, 1991). The paleobotanical results
suggest the following sequence:
Paleocene: after a brief initial low, temperatures began
to rise; precipitation increased.
Early Eocene: temperatures increased to maximum val-
ues; continued high precipitation.
Middle Eocene: temperatures began to decline; rainfall
decreased and became more seasonal.
In an analysis of 15 localities in the Atlantic and Pa-
cific Oceans, values for kaolinite, a hydrous aluminum
silicate indicative of humid conditions, are high at the
K-T boundary and decrease in the Middle Eocene. The
abundance of this and other clay minerals also covaries
with 18O values that were relatively high at the K-T
boundary (cooling from mid-Cretaceous highs), lowest
in the Middle Eocene (warmest climates; kaolinite in-
creases), and higher at the end of the Middle Eocene with
the onset of cooling (kaolinite decreases). Studies are also
underway to model Eocene climates, but at this early
stage some differences still exist between inferred paleo-
climatic information and model results (Sloan and Bar-
ron, 1992; see previous discussion of the Green River
flora). Events that may prove relevant to changes at the
Eocene-Oligocene boundary are the near simultaneous
impacts of asteroids at Chesapeake Bay and in Siberia
(Chapter 2).
OLIGOCENE VEGETATION: 35.4-23.3 MA
There are few recently published studies on plant megafos-
sil or microfossil assemblages of Oligocene age from the
southeastern United States (Fig. 6.13). Megafossils are
known from the Catahoula Formation of eastern Texas

Table 6.10. Alaskan Paleogene floras and inferred climates.

Assemblage Age MAT (°C) CMMT (°C) WMMT (°C) MART (°C) MGSP (cm)

Redoubt E Oligo 9.0 -4.0 22.0 26.0 60?

Rex Creek LEo 12.3 0.3 24.3 24.0 40?
Katalla LEo 16.2 6.6 25.6 19.2 120
Carbon Mt. M-LEo 8.7 -0.8 18.2 19.0 ?
Aniakchak MEo 13.7 3.3 24.1 20.8 >145
Charlotte Ridge MEo 17.2 8.6 25.8 17.2 >145
Kulthieth EEo 19.4 12.6 26.2 13.6 >145
Chickaloon L Paleo 12.3 1.1 23.5 22.4 >145

MAT, mean annual temperature; CMMT, cold-month mean temperature; WMMT, warm-month mean temperature; MART, mean annual range of temper-
ature; MGSP, mean growing season precipitation. Adapted from Wolfe (1994a, p. 230).
 218
218 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 6.13. Oligocene time scale. Reprinted from Berggren et al. (1995) with the permission of the Society for Sedi-
mentary Geology.

(Berry, 1916); however, the preservation is poor, most iden-
tifications have not been confirmed, and the age of the
Catahoula Sandstone within the Oligocene is uncertain.
Some recently collected material includes Eomimosoidea
plumosa (Leguminosae; Daghlian et al., 1980), also known
from the Middle Eocene Claiborne Formation of western
Kentucky-Tennessee, and Quercus (Fagaceae; Daghlian
and Crepet, 1983). The Vicksburgian pollen-spore flora
(Table 6.11) records a continued decline in diversity as
more tropical elements were eliminated from the region
due to cooling temperatures and more seasonal rainfall
(Frederiksen's, 1991 fourth vegetation event). The Early
Oligocene overall was a time of cooling and increased sea-
sonal dryness, corresponding to the first definitive evi-
dence for substantial ice sheets on Antarctica.
Oligocene vegetation in the Plains region is not well
known (Celtis, hackberry, is common), although Retallack
(1990), using evidence from paleosols, believes the vegeta-

Table 6.11. Palynomorphs from Vicksburgian provincial stage (Early Oligocene) of Gulf Coast of United States.
Palynomorphs
Gymnosperms
Cupressacites hiatipites
Monocotyledon Angiosperms
Aglaoreidia pristina
Milfordia sp.
Palmae
Sparganium/Typha type
Dicotyledon Angiosperms
Cupuliferoipollenites sp.
Fraxinoipollenites sp.
Momipites
Myrtaceidites sp.
Quercus?
Siltaria sp.
Striopollenites terasmaei
Adapted from Frederiksen (1981).
tion was scrubland growing under a regime of low rainfall
that was seasonally distributed.
In the west one of the most throughly analyzed megafos-
sil assemblages is the Late Oligocene Creede flora of south-
western Colorado about 180 km southwest of Florissant.
Axelrod (1987) recognized 73 species (19 gymnosperms, 54
angiosperms), and Pinus (10 species) and Ribes (eight
species) were the most diverse genera (Table 6.12). The de-
positional setting was a structural moat surrounding a
caldera (crater). Two major communities were recognized:
a mixed hardwood-coniferous forest and a pinon-juniper
woodland on exposed drier and warmer slopes. (Modern
analog species are Pinus edulis and Juniperus osteo-
sperma.} Other elements record a riparian community and
aquatic vegetation. The paleoelevation was placed near
1200-1400 m, and the walls of the caldera rose to a nearby
volcanic plateau at 2100-2400 m. Annual precipitation
was estimated at 460-635 mm (summer wet), the MAT at
11.5°C (1.9°C today), and the annual range of temperature
at 16°C (25°C today). Notable features of this higher ele-
vation mixed coniferous forest and drier pifion-juniper
woodland scrub were the rarity of present-day mid- to
lower elevation mesic Asian or eastern North American ex-
otics that characterize other western Oligocene and Mio-
cene floras and the small size of the leaves. This was attrib-
uted to reduced summer rain enhanced by the local
physiographic setting. Affinities of the Creede flora were
cited as primarily with the local modern vegetation and
that of the southern Rocky Mountains of Arizona, New
Mexico, and northwestern Mexico (the Madrean region).
The area is an important center of diversification for Pinus
Middle Eocene through Early Miocene History 219
Affinities
Taxodiaceae-Cupressaceae-Taxaceae
Unknown
Unknown
Sparganiaceae -Typhaceae
Fagaceae
Unknown
Juglandaceae, Engelhardia group
Unknown
Fagaceae
Fagaceae
Unknown
and Quercus. With the development of cold winters and
drier summers in the southern Rocky Mountains in
Miocene and later times, many of the pines and oaks were
eliminated there and are now found in the Sierra Madre
Occidental of Mexico.
In a restudy of the Creede flora (dated at 27.2 Ma incor-
porating new decay constants), and using mostly the same
collections, only 34 taxa were recognized (Wolfe and
Schorn, 1989, 1990; Table 6.12). As would be expected
from these taxonomic treatments, the paleoenvironmental
reconstructions are also different. The latter study pro-
poses four major communities: fir-spruce forest, fir-pine
forest, pine-juniper forest or woodland, and mountain
mahogany (Cercocarpus) chaparral, with several species
having affinities with eastern North America and eastern
Asia. The associations are considered distinct from similar
modern communtities, and the Creede forests are de-
scribed as having no exact modern analog. The MAT de-
rived from the CLAMP program is ~4.5°C (4.2°C ; Wolfe,
1993), which is fully 7°C cooler than the estimate by Axel-
rod (1987).
Among other things, these two taxonomic treatments il-
lustrate the latitude that still exists in the identification of
paleobotanical material. The revisions provided by Wolfe
and Schorn (1990) are followed here to characterize the
vegetation during Creede time. One generalization pro-
vided by both studies is that the montane coniferous forest,
which was evident early in the Princeton and Republic flo-
ras, and drier shrubland-woodland vegetation present to
the east of the Rocky Mountains (e.g., in the Green River
flora) continued to develop from precursors that were pres-
Table 6.12. Megafossils from Late Oligocene Creede flora, Colorado.
Axelrod (1987)

Gymnosperms
Cupressaceae Pinaceae con't.
Juniperus creedensis Pinus crossii
J. gracillensis P. engelmannoides
Pinaceae P. florissantii
Abies concoloroides P. macginitieii
Aoides p.ponderedoies
Picea coloradensis P. riogrande
P. lahontensis P. sanjuanensis
P. sonomensis P. wasonii
Pinus alvordensis Pseudotsuga glaucocoides
P. coloradensis Tsuga pentranensis

Monocotyledon Angiosperms
Cyperaceae
Cyperacites creedensis

Dicotyledon Angiosperms
Aceraceae Ranunculaceae
Acer riogrande Ranunculus creedensis
Betulaceae Rhamnaceae
Betula smithiana Condalia mohavensis
Berberidaceae Zizyphus florissantii
Berberis coloradensis Rosaceae
B. riogrande Cercocarpus holmesii
Mahonia creedensis C. linearifolius
M. obliqua Chamaebatiaria creedensis
Bignoniaceae Crataegus creedensis
Chilopsis coloradensis Fallugia lipmanii
Caprifoliaceae Holodiscus harneyensis
Symphoricarpos wassukana H. idahoensis
Elaeagnaceae Peraphyllum septentrionale
Shepherdia creedensis Physocarpus petiolaris
Ericaceae P P. triloba
Arbutus stewartii Prunus chaneyii
Fagaceae P. creedensis
Quercus creedensis Rosa hilliae
Grossulariaceae Rubus riogrande
Ribes birdseyii Sorbus potentilloides
R. creedensis Salicaceae
R. dissecta Populus cedrusensis
R. lacustroides P. creedensis
R. riogrande P. pliotremuloides
R. stevenii Salix creedensis
R. wasonii S. venosiuscula
R. webbii Sambucaceae
Hippuridaceae Sambucus longifolius
Hippurus coloradensis Sapindaceae
Leguminosae Sapindus coloradensis
Cercis buchananensis Saxifragaceae
Robinia californica Fendlera coloradensis
Nymphaeaceae famesia caplanii
Nuphar coloradensis Philadelphus creedensis
Oleaceae Vacciniaceae
Fraxinus creedensis Vaccinium creedensis

Table 6.12. continued
Wolfe and Schorn (1990)
Abies rigida
Picea magna
Pinus crossii
P. sanjuanensis
P. riogrande
P. sp. 1
P. sp. 2
Juniperus creedensis
Berberis coloradensis
Mahonia aceroides
Dilleniaceae
Populus larsenii
Salix sp.
famesia caplanii
Ribes lacustroides
R. obovatum
R. robinsonii
Eleiosina praeconcinna
Adapted from Axelrod (1987) and Wolfe and Schorn (1990).
ent in the Middle Eocene. Subtropical elements were es-
sentially eliminated from the central Rocky Mountains in
the Oligocene, and overall the forests became more modern
in aspect during this time (Leopold and MacGinitie, 1972).
To the west in north-central Oregon Early Oligocene
plants are known from the John Day Formation in the
Crooked River and John Day basins in the same vicinity as
the Late Eocene Clarno flora. The classic locality at Bridge
Creek is Early Oligocene and has been radiometrically
dated at 31.8 and 32.3 Ma (Evernden et al, 1964; corrected
to new constant; more recently 32.24-33.62 Ma, Meyer
and Manchester, 1996); therefore, it is younger than the
Clarno (44-43 Ma). The Bridge Creek flora was studied by
Chaney (1924,1927), individual elements are part of wider
revisions (Brown, 1939, 1946, 1959; Manchester, 1992,
Florissantia speirii; Manchester and Crane, 1987; Tanai
and Wolfe, 1977; Wolfe, 1977; Wolfe and Tanai, 1987), and
there is a recent revision by Meyer and Manchester (1997;
91 genera, 110 species). Current studies are focused on a
locality at Fossil, Oregon (32 Ma, SCLF technique), just to
the northeast of Clarno, and about 35 mi. north of the prin-
cipal Bridge Creek locality. Manchester and Meyer (1987)
presently recognize 35 species from the lacustrine shales at
fossil (Table 6.13), including Metasequoia, Acer, Alnus,
Paracarpinus (leaves), Asterocarpinus (extinct, fruit; Man-
chester and Crane, 1987), and Pteleaecarpum (Craigia; see
Clarno flora) as the most abundant. Eucommia occurs at
other localities in the John Day Formation (Call and Dilcher,
1997). Of particular interest is Cercidiphyllum (katsura,
eastern Asia), which was widespread in mid-Tertiary floras
across the northern hemisphere, because the Early Oligo-
cene occurrence in the Bridge Creek flora is among the old-
est known; earlier similar forms are now assigned to re-
Middle Eocene through Early Miocene History 221
Stockeya creedensis
Holodiscus stevenii
Crataegus creedensis
Sorbus potentilloides
PotentiUa creedensis
Cercocarpus hendricksonii
C. nanophyllus
Osmaroniai stewartiae
Prunus creedensis
P. sp.
Rosoideae
Cercis sp.
Legume
Catalpa coloradensis
Eleopoldia lipmanii
Monocotylophyllum sp.
lated but extinct genera (Crane and Stockey, 1985). The
vegetation is principally a temperate broad-leaved decidu-
ous hardwood forest with many elements similar to those
in the mixed northern hardwood forest of North America
and eastern Asia. The MAT is estimated at 9-ll°C with a
mean annual range of ~23°C. Broad-leaved evergreens are
rare. Compared to the Late Eocene Clarno flora in the same
area (predominantly broad-leaved evergreen), the Early
Oligocene Bridge Creek-Fossil floras (predominantly
broad-leaved deciduous) reflect a trend toward a cooler,
drier, more seasonal climate.
The Rujada flora of west-central Oregon is assigned to
the Late Oligocene by Lakhanpal (1958) and to the Early
Oligocene by Wolfe (1981b). It preserves remnants of the
vegetation growing along the shores of Oregon adjacent to
the rising Cascade Mountains. The name derives whimsi-
cally from the initials of two loggers (R. Upton and J. Ander-
son) and a forestry project undertaken by the Department of
Agriculture. The flora preserves a familiar assemblage of
mesic deciduous hardwood elements and higher altitude
gymnosperms with affinities to the modern vegetation of
eastern North America, western North America, and east-
ern Asia. The presence of a few warm-temperate to sub-
tropical elements and broad-leaved evergreens (Annona,
Machilus, Nectandra, Ocoted], along with some wider
ranging plants also found in warm-temperate vegetation
(Alangium, Engelhardia), reflect the maritime environ-
ments of the coast rather than the increasingly dry climates
of the intermountain basins and the south-facing or lee-
ward slopes to the west. The presence of broad-leaved de-
ciduous, broad-leaved evergreen, and coniferous elements
is similar to the mixed broad-leaved evergreen and conifer-
ous forests of eastern Asia. The MAT is estimated at
222 Late Cretaceous and Cenozoic History of North American Vegetation

Table 6.13. Megafossils from John Day Formation, Fossil, Oregon.

Pteridophytes
Cf. Polypodium
Gymnosperms
Pinaceae Taxodiaceae
Pinus sp. Metasequoia occidentalis
Abies sp.
Monocotyledon Angiosperms
Monocotyledonous leaf
Dicotyledon Angiosperms
Aceraceae Leguminosae
Acer ashwilli Cercis sp.
A. cranei Cladrastis sp.
A. manchesteri Meliaceae
A. osmonti Cedrela merilli
Berberidaceae Moraceae
Mahonia simplex Morus-like leaves
Betulaceae Platanaceae
Alnus hollandiana Platanaceous fruits
Asterocarpinus perplexans Platanus aspera
Bombacaceae -Tiliaceae - Sterculiaceae P. condoni
Florissantia speirii Rosaceae
Cercidiphyllaceae Crataegus newberryi
Cercidiphyllum crenatum Cf. Sorbus
Fagaceae Tiliaceae
Fagus pacifica Plafkeria obliquifolia
Quercus consimilis Pteleaecarpum (Craigia)
Hydrangeaceae oregonense
Hydrangea florissantia Tilia circularis
Juglandaceae Ulmaceae
Cruciptera Tremophyllum hesperium
Cf. Engelhardia olsoni Ulmus pseudo-americana
Juglans sp.
Pterocarya sp.
Adapted from Buzek et al. (1989), Kvacek et al. (1991), Manchester (1991, 1992), and Manchester and Meyer (1987).

12-13°C. The presence of Abies suggests that the Cascade
Mountains, having reached an elevation to support this
high-elevation conifer, were likely beginning to cast a rain-
shadow locally across the adjacent interior basins (e.g., the
drier Bridge Creek and Fossil floras). The Late Oligocene
Yaquina flora of western Oregon (McClammer, 1978) is pre-
served in delta deposits and represents a humid to mesic
coastal rain forest. The CLAMP estimate of MAT is 15.5°C
(Wolfe, 1993), and some adjustment in interpretation is
necessary between the floras at Rujada (age uncertain,
MAT based on modern analogs) and Yaquina (MAT based
on CLAMP).
To the far northwest, development of temperate decidu-
ous and coniferous forests was favored by the Middle
Eocene through Early Oligocene cooling. Palynomorphs
from the lower part of the Richards Formation in the
Mackenzie Delta region of northern Canada record both
cool (Sphagnum, Picea, Alnus, Betula) and warmer (Meta-
sequoia, Pinus, Sequoia, Tsuga, Castanea, Pterocarya, Quer-
cus) elements in the Late Eocene (Norris, 1982), likely re-
sulting from wind transport of pollen from higher altitude
vegetation into the lowlands. In the upper part of the for-
mation (Early and Middle Oligocene) a decrease in diver-
sity and a change in composition of the assemblage is evi-
dent. Because the change is not associated with facies
differences, it is likely attributable to climate. The common
palynomorphs are Sphagnum, Lycopodium, Osmunda,
Picea, Pinus, Tsuga, Alnus, Betula, Carpinus, and Ptero-
carya; the more thermophylic Sequoia, Castanea, Quercus,
Tilia, and Ulmus disappear. [Compare these palynomorph
assemblages with elements of the modern closed forest
phase of the boreal coniferous forest formation and the de-
ciduous forest formation of the southern Appalachians
(Chapter 2).] These return later in the Oligocene, and the
changes in vegetation probably reflect the same fluctua-
tions previously discussed that occurred during the Mid-
dle Eocene through the Oligocene in the southeastern United
States and southern California and that are paralleled in
 Middle Eocene through Early Miocene History 223

the marine-based paleotemperature record (Fig. 3.1). In Zanthoxylum; Tiffney, 1980), Theaceae (cf. Cleyera], Sar-
Alaska the Salicaceae (Populus, Salix) diversified during gentodoxa (Sargentodoxaceae, a deciduous vine presently
the Eocene-Oligocene temperature decline to become the restricted to southeast Asia; Tiffney, 1993), Turpinia
dominant streamside element and then spread southward (Staphyleaceae; Tiffney, 1979), and Vitaceae (Tiffney and
to characterize midnorthern latitude gallery forests by the Barghoorn, 1976). A complete listing of the megafossils
Early Miocene. This characteristic riverine assemblage first identified from the Brandon flora is given in Tiffney
appears as a well-defined community in the mid- to Late (1994). The habitat was river swamps similar to those of
Eocene of the high northern latitudes. the mesothermal Atlantic-Gulf Coastal Plain further to
The reassignment in age of several North American flo- the south, as indicated by Cyrilla, Gordonia, Persea, and
ras has affected the data base available for interpreting Symplocos, which presently do not extend northward
Oligocene vegetation and terrestrial environments. The into areas with significant frost. The climate was warm-
placement of the Brandon lignite flora from the Oligocene temperate to subtropical, essentially frost free (MAT 17°C;
to the Early Miocene, and the Bridge Creek, Fossil, Floris- presently 7.6°C) with -1660 mm of annual precipitation
sant, Beaverhead Basin, Metzel Ranch, York Ranch, Mor- (presently 950 mm; Tiffney, 1993). Present-day Asian gen-
mon Creek, Ruby, and other floras from the Oligocene to era such as Glyptostrobus, Alangium, and Pterocarpus
the Late Eocene—Early Oligocene leaves few assemblages were present, as in other mid-Tertiary floras across the
for the period between -36-23 Ma. This emphasizes the northern hemisphere. In New England the vegetation had
importance of the Creede flora, the need to explore for changed from the tropical forest of the Early Eocene to de-
other floras of Oligocene age, and the enhanced role of fau- ciduous broad-leaved or mixed mesophytic forest by the
nas and context information in paleoenvironmental recon- Early Miocene.
structions (Figs. 3.1, 3.13, 6.1). A challenging problem has been to determine the age of
the lignite that is underlain by the Lower Cambrian Cheshire
Quartzite, overlain by Quaternary drift, and lacks radio-
EARLY MIOCENE VEGETATION: 23.3-16.3 MA metrically datable material and vertebrate remains. An in-
novative approach was developed by Barghoorn (1950,
The climatic patterns of the Oligocene, induced in part by 1951, 1953) and Spackman (1949) based on Reid (1920).
early continental glaciations on Antarctica, continued Percentages of exotic genera in Tertiary floras of known age
into the Early Miocene but with a slight warming trend were plotted against time, and the resulting curve was used
that is evident until about 15 Ma (Figs. 3.1, 6.14). Re- to estimate the age of various floras. In the case of the Bran-
cently studied Early Miocene floras are not widely avail- don Lignite this value was 72%, which corresponded to an
able from the southeastern United States, but the existing age of Late Oligocene (Traverse, 1955, fig. 5). Other age esti-
information suggests continued warm-temperate to sub- mates are Eocene-Oligocene to earliest Miocene (35.4-
tropical conditions. 23.3 Ma; Tiffney, 1993), but the most recent assignment is
To the northeast few Tertiary floras are known from this to the Early Miocene (Tiffney, 1994; Tiffney and Traverse,
mostly Paleozoic and Mesozoic province. An exception is 1994; Traverse, 1994).
the Brandon Lignite in west-central Vermont that was dis- Shortly after publication of the age-area curve, Axelrod
covered in 1848 during mining operations and was subse- (1957) published a critique to which Wolfe and Barghoorn
quently used to generate steam for the Brandon Iron and (1960) responded. The arguments presented in support of
Car Wheel Company. The lignite has long been known for these and other often diametrically opposed conclusions
its fossils, which include pollen, spores, other kinds of mi- (e.g., Axelrod, 1952,1961; Scott et al., 1960) reflect the de-
crofossils (e.g., colonial algae), fruits, seeds, and wood. bated nature of many aspects of vegetational history. In the
Palynomorphs from the Brandon Lignite identified as to meantime, considerable new information, innovative ideas,
biological affinities by Traverse (1955) are listed in Table and revisions are generated and progress is made. In retro-
6.14 (see also Traverse, 1994). The most abundant taxa are spect, the age-area curve gave an age estimate (Late Oligo-
Carya, Castanea, Coiy/us?, Cyrilla, Engelhardia, Gordonia, cene) that was near the most recent one (Early Miocene).
Ilex, Manilkara, Morus, Nyssa, Planera, Quercus, Rham- To the far north on Devon Island, Northwest Territories
nus, Siltaria (extinct genus), Ulmus, and Vitis. Megafossils (75° 22' N, 89° 40' W), lacustrine sediments of the Haugh-
(mostly fruits and seeds) are under study and presently in- ton Formation from an impact crater 16 km wide at 22.4 ±
clude Alangium (along with pollen, Fig. 6.15; Eyde et al., 1.4 to 23.4 ± 1 Ma (Early Miocene, Aquitanian; Omar et al.,
1969), Caldesia (Alismataceae; Haggard and Tiffney, 1997; 1987) contain plant micro- and megafossils and faunal re-
also in the Clarkia flora of Idaho), Cyperaceae (B. H. mains of a mixed coniferous-hardwood assemblage domi-
Tiffney, personal communication, 1996), Caricoidea (ex- nated by Larix, Picea, Pinus, Tsuga, Alnus, Betula, Corylus,
tinct, monocotyledonae), Illiciaceae (Tiffney and Barg- Ericales, and Ulmus-Zelkova (Table 6.15). The closest
hoorn, 1979), Magnoliaceae (Tiffney, 1977), Microdiptera modern analog is the cool-temperate conifer-hardwood
(Tiffney, 1981), Moraceae (Moroidea), Nyssa (Eyde and vegetation transitional between the boreal forest and the
Barghoorn, 1963; see also Eyde, 1997), Rutaceae (Euodia, deciduous forest in northeastern maritime North America.
 h

Figure 6.14. Early Miocene time scale. Reprinted from Berggren et al. (1995) with the permission of the Society for Sedimentary Geology.
Table 6.14. Fossils from Early Miocene Brandon Lignite, Vermont.

Algae
Chlorophyta Dinoflagellates
Botryococcus Peridinium

Pteridophytes
Fern spores (unidentified)

Gymnosperms
Pinaceae Taxodiaceae
Pinus (haploxylon type) Glyptostrobus
Pinus (sylvestris type)

Monocotyledon Angiosperms
Gramineae

Dicotyledon Angiosperms
Alangiaceae Moraceae
Alangium Morus
Anacardiaceae Nyssaceae
Rhus Nyssa
Aquifoliaceae Onagraceae
Ilex Jussiaea
Betulaceae Rhamnaceae
Corylus! Rhamnus
Clethraceae Rosaceae
ClethraW Rubus (x)
Ericaceae Rosaceae(?)
Ericaceae (x) Rutaceae
LyoniaC?) Horniella (extinct genus)
Oxydendrum(?) Phellodendron (x)
Rhododendron Toddalieae
Vaccinium Santalaceae
Zenobia (x) Nestronia(t)
Fagaceae Sapotaceae
Castanea Manilkara
Fagus Mimusops
Quercus Symplocaceae
Siltaria (extinct genus) Symplocos
Hamamelidaceae Theaceae
Liquidambar Gordonia
Juglandaceae Tiliaceae
Carya Tilia
Engelhardia Ulmaceae
Juglans Ulmus
Pterocarya Planera
Lauraceae Vitaceae
Cinnamomum-iype (x) Vitis (Subg. Muscadinia group (x)
Ocotea-type (x)
Persea (x)
Magnoliaceae
IHicium (x)
Magnolia (x)

Adapted from Traverse (1955); x, megafossil only). See Traverse (1994) for list of palynomorphs assigned to form taxa and text for other megafossils
described later.

225
226 Late Cretaceous and Cenozoic History of North American Vegetation

Table 6.15. Pollen and spores from Early Miocene

Haughton Formation, Devon Island, Arctic Canada.
Bryophyta Dicotyledon Angiosperms
Sphagnum type Acer
Pteridophytes Alnus
Botrychium type Betula
Figure 6.15. Pollen of Dryopteris type Carya
Alangium barghoorni- Lycopodium type Castanea
anum, Brandon Lig- Osmunda type Chenopodiineae
nite .Vermont. Adapted Pteridium type Corylus type
from Traverse (1955). Selaginella Cruciferae
Gymnosperms Ericales
Abies Cf. Fagus
Cupressaceae Cf. Fraxinus
Larix type Ilex
The MAT estimates range widely between 0 and 11°C Picea Juglans
(Hickey et al., 1988) and 15-20°C (Whitlock and Dawson, Pinus Liquidambar
1990). Eliminating the high and low extremes, the MAT fig- Pinus strobus type Ostrya-Carpin us
ures are 11-15°C (compare, e.g., 17°C for the Brandon lig- Tsuga Populus
nite). In the Canadian Maritime Provinces precipitation is Monocotyldeon Angiosperms Pterocarya
Cyperaceae Quercus
-1000 mm/year. Potamogeton Salix
To the west on Banks Island, a flora of late Early to early Sparganium type Ulmus- Zelkova
Middle (Seldovian) Miocene age is preserved in the Mary
Adapted from Whitlock and Dawson (1990).
Sachs Gravel. This informal unit was named to differenti-
ate it from the Beaufort Formation (type locality Prince Pat-
rick Island) that, through imprecise usage, had come to in-
clude unrelated deposits with a wide range of ages. The In the central Rocky Mountains tectonic activity had
flora is one of the richest in the Neogene of Arctic North waned in the Middle Tertiary, but in the Miocene it re-
America (Matthews and Ovenden, 1990; Table 6.16); it also sumed. Erosion from uplift combined with volcanism de-
reflects mixed coniferous-hardwood forest having affini- posited extensive sandstones and tuff (water-lain volcanic
ties with eastern Asia (Glyptostrobus, Metasequoia), east- ash). In Jackson Hole, Wyoming, up to 2300 m (7000 ft) of
ern (including southeastern) North America (Phyllanthus), volcanics were deposited, and in the San Juan Mountains
and extant local vegetation (Abies, Thuja, Alnus, various of southern Colorado an area of 25,000 km2 was filled with
Ericaceae, Menyanthes}. In addition to the taxa listed in 20,000 km3 of ash and tuff. The climatic history for the
Table 6.16, others from smaller floras in the region include northern Rocky Mountains is sketchy because few Neo-
Chara-Nitella type, cf. Tubela, Nuphar, Nymphaea, cf. gene floras are available. The region was likely a western
Vitis, Viola sp., Decodon sp., Hippuris sp., Vaccinium sp., montane coniferous forest with an understory and stream-
Nymphoides, and Lycopus. In some sections of the Beau- side component of broad-leaved deciduous angiosperms;
fort Formation younger than 22 Ma (Banks Island) and ~3 but as emphasized previously, apparent stasis is usually an
Ma (Meighen Island), nonarboreal taxa are common artifact of poor resolution. Sedimentary patterns, paleosol
(Gramineae, Artemisia, various herbs). In one section from mineralogy, and vertebrate assemblages suggest fluctua-
Meighen Island an early and local forest-tundra vegeta- tions in precipitation: wet to dry in the Late Eocene, dry to
tion was present (Matthews, 1987). Tree taxa are fewer in wet in the late Early Miocene, wet to dry in the early Mid-
the Late Pliocene-Early Pleistocene (Matuyama age) floras dle Miocene, and dry to wet in the Late Miocene-Pliocene
of the Kap K0benhavn Formation. These floras record the transition (Thompson et al., 1982). These fluctuations con-
development of tundra from ~3 Ma to the present and are tributed to the dynamic nature of the vegetation and un-
discussed in Chapter 7. doubtedly to evolutionary diversification.
Little is known of the Early Miocene vegetation in the In the western Great Basin, vegetation near the Early to
Plains area. A shrubland-savanna likely occupied the re- Middle Miocene boundary is preserved in the Buffalo
gion, and a few grass fossils representing extinct genera Canyon flora of western Nevada (15.6 Ma; Axelrod, 1991;
begin to appear. These persist into the Middle and Late Wolfe et al., 1997; Table 6.17). The largest families are
Miocene when they are augmented by additional grass gen- Pinaceae (10 species), Rosaceae (11), Salicaceae (10), Acer-
era (Stipidae and Poidae) and prairie herbs (e.g., Boragi- aceae (5), and Betulaceae (5). The largest genera are Salix (7
naceae). A small flora from the Marsland Formation in species), Acer (5), Betula (4), Populus (4), Ribes (4), Picea
northwest Nebraska (19-18 Ma; Hemingfordian NALMA) (3), and Prunus (3). Most species presently occur in the
includes Acer cf. negundo (box elder), Crataegus (haw- western United States, and a few occur in the eastern United
thorn), Robinia (locust), and Ulmus (elm). Annual rainfall States and eastern Asia. It is a mixed conifer-deciduous
is estimated at 900 mm (Axelrod, 1985a). hardwood forest with representatives of the adjacent
Table 6.16. Plant megafossils from Mary Sachs Gravel southern Banks Island, N.W.T.

Gymnosperms
Cupressaceae Pinaceae cont.
cf. Thuja occidentalis P. palaeodensiflora
Pinaceae cf. P. funebris
cf. Abies grandis Tsuga sp.
cf. Larix omoloica Taxodiaceae
Larix sp. Glyptostrobus sp.
Picea banksii Metasequoia disticha
P. sp. M. sp.
Pinus five-needle type Taxodium sp.
P. itelmenorum

Monocotyledon Angiosperms
Alismataceae Potamogetonaceae
cf. Sagisma cf. Potamogeton richardsonii
Araceae Sparganiaceae
Epipremnum crassum Sparganium sp.
Cyperaceae
Carex sp.

Dicotyledon Angiosperms
Actinidiaceae Juglandaceae
Actinidia sp. Juglans eocineria
Aizoaceae Labiateae
cf. Sesuvium Teucrium sp.
Araliaceae Lythraceae
Aralia sp. Microdiptera/Mneme type
Betulaceae Magnoliaceae
Alnus (Alnobetula) sp. Liriodendron sp.
Cf. Alnus incana Moraceae
Cf. Betula apoda Morus sp.
Betula dwarf shrub type Myricaceae
B. arboreal type Comptonia sp.
Capparidaceae Myrica (Gale) sp.
Cleome sp. Onagraceae
Polanisia sp. Ludwigia sp.
Caprifoliaceae Polygonaceae
Diervilla sp. Rumex sp.
Sambucus sp. Ranunculaceae
Weigela sp. Ranunculus (Batrachium) sp.
Chenopodiaceae R hyperboreus
Chenopodium sp. Rhamnaceae
Crassulaceae Cf. Paliurus sp.
Sedum type Rosaceae
Ericaceae Potentilla sp.
Andromeda polifolia Rubus sp.
Arctostaphylos alpina/rubra type Saxifragaceae
Chamaedaphne sp. Genus?
Euphorbiaceae Solanaceae
Phyllanthus (Phyllanthus) sp. Solanum/Physalis type
Gentianaceae Verbenaceae
Menyanthes small form Verbena sp.
Hypericaceae
Hypericum sp.
Adapted from Matthews and Ovenden (1990).
Table 6.1 7. Plant megafossils from Early-Middle Miocene Buffalo Canyon flora, western Nevada.

Gymnosperms
Cupressaceae
Chamaecyparis cordillerae
Juniperus desatoyana
Pinaceae Pinaceae cont.
Abies concoloroides P. sonomensis
A. laticarpus Pinus balfouroides
Larix churchillensis P. ponderosoides
Picea lahontense Pseudotsuga sonomensis
P. magna Tsuga mertensioides

Monocotyledon Angiosperms
Typhaceae
Typha lesquereuxii

Dicotyledon Angiosperms
Aceraceae Meliaceae
Acer median um Cedrela trainii
A. negundoides Myrtaceae
A. oregonianum Eugenia nevadensis
A. trainii Nymphaeaceae
A. tyrrellii Nymphaeites nevadensis
Berberidaceae Oleaceae
Mahonia macginitiei Fraxinus desatoyana
M. reticulata F. eastgatensis
Betulaceae Rosaceae
Alnus latahensis Amelanchier desatoyana
Betula ashleyii Cercocarpus ovatifolius
B. desatoyana Chamaebatia nevadensis
B. idahoensis Crataegus middlegatei
B. thor Heteromeles desatoyana
Caprifoliaceae Lyonothamnus parvifolius
Symphoricarpos wassukana Prunus chaneyii
Carpinaceae P. moragensis
Carpinus oregonensis P. treasheri
Ericaceae Rosa harneyana
Arbutus trainii Sorbus cassiana
Fabaceae (Leguminosae) Salicaceae
Amorpha stenophylla Populus cedrusensis
Robinia bisonensis P. eotremuloides
Fagaceae P. payettensis
Chrysolepis sonomensis P. pliotremuloides
Quercus hannibalii Salix churchillensis
Q. wislizenoides S. desatoyana
Grossulariaceae S. laevigatoides
Ribes barrowsae S. owyheeana
R. bonhamii S. pelviga
R. stanfordianum S. storeyana
R. webbii Ulmaceae
Hydrangeaceae Ulmus speciosa
Hydrangea bendirei Zelkova brownii
Juglandaceae Vacciniaceae
Carya bendirei Vaccinium sophoroides
Juglans desatoyana
Adapted from Axelrod (1991).

broad-leaved evergreen sclerophyll vegetation from lower
elevations on warmer and drier sites. The MAT is esti-
mated at 10°C (CLAMP 8.7°C), mean annual range at 14°C,
and annual precipitation at 900-1000 mm (35-40 inches)
with ample summer rain. The paleoelevation of the lake
basin was estimated at -1280 m (4200 ft; presently 1900
m). However, many of the Early and Middle Miocene floras
of the Basin and Range Province may have been consider-
ably higher based on leaf physiognomy and mean annual
enthalpy calculations (e.g., Buffalo Canyon, 3.2 km; East-
gate, 2.7 km; Middlegate, 2.8 km; Wolfe et al., 1997, table
1). Axelrod (1985b) attributes differences between the Mid-
dlegate and Eastgate floras (15.5 Ma) to slope exposure.
The Middlegate flora is dominated by sclerophyllous trees
and shrubs (shrubland-chaparral) living on drier south-
facing slopes. The vegetation included Arbutus prexala-
pensis, Cedrela trainii, Cercocarpus antiquus, C. pacifica,
Lithocarpus nevadensis, Quercus hannibali, Q. shrevoides,
and Q. simulata. The MAT is estimated at 10.2°C
(CLAMP). The Eastgate flora has a better representation of
mixed coniferous-hardwood species from north-facing
canyons (Abies concoloroides, Larix cassiana, L. nevaden-
sis, Amelanchier grayi, Aesculus preglabra). The MAT is
estimated at 10.3°C (CLAMP).
To the north the Alvord Creek flora (-21 [21-23] Ma,
southeastern Oregon; Axelrod, 1944) is also an upland
coniferous-hardwood forest. To the south the Tehachapi
flora (17.5 [16.1] Ma; southeastern end of the Sierra
Nevada; Axelrod, 1939) is an evergreen oak-cypress-
pinon woodland and associated arid subtropical scrub
growing at lower elevations in a warmer climate. To the
west, the Sutro flora (21 Ma) near Silver City, Nevada, is a
richer mesophytic exotic broad-leaved evergreen and
conifer vegetation on the windward side of the low Sierra
Nevada (Axelrod, 1991).
In the far northwest the Seldovia Point flora from the
Kenai Group, Cook Inlet region of Alaska, is Late Seldovian
in age (late Early to early Middle Miocene, -15.8-16.8
[14.5-14.2] Ma; Fig. 6.11). As with the vegetation to the
east across the high northern latitudes, it represents a
mixed mesophytic hardwood forest of Celtis, Fagus,
Juglans, Liquidambar, Magnolia, Nyssa, Platanus, Ptero-
carya, and Zelkova (megafossil evidence), with coniferous
forests close by (microfossil evidence; Wolfe and Tanai,
1980). The MAT is estimated at 6-7°C (CLAMP 9°C) with a
mean annual range of ~26°C . A summary of MAT and
MART for western Washington and Oregon for the Late
Eocene through the Miocene is given in Fig. 6.16. It shows
the dramatic increases in MART across the Eocene-
Oligocene boundary.
As shown on the global benthic paleotemperature curve
(Fig. 3.1), the Earth's biotas experienced another downshift
in temperature and other environmental changes in the
Middle Miocene (15-14 Ma) that ushered in a new round of
evolutionary activity, a reshuffling of distributions, and con-
tinued modernization of the North American vegetation.
Middle Eocene through Early Miocene History 229
VEGETATION SUMMARY
During the Middle Eocene the southeastern United States
was occupied by a semideciduous tropical dry forest (Table
6.18). Mangrove vegetation of Nipa and associated Acros-
tichum fringed the coast. Nipa disappeared at the end of
the Eocene and the brackish-water habitats eventually be-
came occupied by elements of the modern neotropical veg-
etation: Avicennia nitida, Laguncularia mcemosa, and Rhi-
zophora mangle associated with Conocarpus erecta. On
sandy sites Pinus and palmetto-like palms (Sabal-Serenoa
type) were present and represent an early version of the
modern sand pine scrub of the pine woods association.
Warm-temperate deciduous elements of Fagaceae and Jug-
landaceae occurred with more tropical elements in the low
to midaltitudes and graded into the cooler temperate vege-
tation in the Appalachian highlands (Picea, Tsuga). The
drying and cooling trend in the Middle and Late Eocene
(Fig. 6.10, 6.16) that supported dry deciduous forest later
became too cool, and moisture increased in the southeast
with redirected pressure systems derived, in part, from
continued rise of the western mountains. This gradually
eliminated the tropical dry forest, favoring the develop-
ment and expansion of warm-temperate deciduous vegeta-
tion at lower altitudes, and the Appalachian montane
coniferous forest in the highlands. Components of associa-
tions of the deciduous forest formation (beech-maple;
oak-chestnut, oak-hickory; southern mixed hardwood,
flood-plain forest, Nyssa, Taxodium, early maple(?)-bass-
wood) in the eastern United States were present in the
Claiborne-Jackson floras; their beginnings, along with the
various edaphic and fire-controlled pine woodlands, can
be traced to the Middle Eocene. Picea is present in the Late
Eocene Jackson flora. The broad sequence of communities
in the southeastern United States has been a Late Creta-
ceous community termed a tropical forest with no exact
modern analog; a Paleocene through Early Eocene tropical
rain forest; a Middle Eocene tropical dry forest; and with
subsequent decline of the dry tropical components, the ap-
pearance, diversification, and expansion of various associ-
ations comprising the modern deciduous forest formation
and the modern Appalachian coniferous forest formation.
Temperature gradients were becoming steeper but were
still lower in the Middle Eocene through the Early Miocene
than at present. Many elements of a warm-temperate broad-
leaved deciduous forest, including present-day Asian ex-
otics, extended through the New England area (Brandon
Lignite flora) and continued as far north as Ellesmere Is-
land (81° N latitude). Acer, Carya, Corylus-Carpinus, Casta-
nea, and Fagus, along with cooler temperate mixed conifer-
ous forest species of Larix, Picea, Tsuga, and Betula formed
a version of the lake states-Canadian maritime forest.
The semideciduous tropical dry forest of the southeast-
ern United States extended northwest around the remnant
of the Cretaceous-Paleocene sea at low to midelevations,
across the Plains, along the eastern side of the Rocky
230 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 6.16. Estimated average annual temperatures (circles) and range of temperatures (lines) for western Oregon and
western Washington based on Cenozoic floras. Localities (1-5) Puget Sound Group; (6-11) near Salem, Oregon; (6)
Bridge Creek, Willamette; (7) Yaquina; (8) Eagle Creek; (9) Latah and other sites; (10) Faraday; (11) Troutdale. Reprinted
from Leopold et al. (1992; based on Wolfe 1978,1981b, 1992a) with the permission of Princeton University Press.

Mountains, and into southern California (San Diego micro-
fossil floras). In the Plains and along the eastern Rocky
Mountains the general sequence of vegetation was from an
Early Eocene moist subtropical community with abundant
paleotropical elements (Wind River flora), a drier subtropi-
cal phase in the Middle Eocene with transition to one with
greater neotropical affinities (Little Mountain flora), a dry
cooling period with a decrease in tropical and subtropical
species (Green River flora), and a Late Eocene phase when
the vegetation was subhumid temperate.
It is becoming clear from the evidence surveyed to this
point that changes in climate proceed in stepwise fashion
and each new level ushers in some modification of the ex-
isting vegetation. Such events allow the perpetuation of re-
stricted, shifting communities growing in edaphically or
slope-controlled environments (e.g., subhumid elements
within an overall more mesic environment). Eventually a
point is reached, however, where further changes cause not
just a modification of existing vegetation, but cross a
threshold wherein the region can no longer support the for-
mation. Such changes may be accelerated, slowed, or tem-
porarily redirected by interplay of the myriad factors
shown at the end of Chapter 2. Nonetheless, if the overall
environmental trend is directional, even though inter-
rupted, a major change in regional vegetation will occur.
This was the case in the the Middle Eocene, and especially
in the Oligocene through the Early Miocene, when forested
areas in the Plains region gave way to a community of
smaller trees, shrubs, herbs, and an increasing number of
grasses, with arborescent forms confined mostly to riverine
habitats. During the Oligocene the canopy became more
open and grasses increased to approach a savanna in the
modern sense. The grass-herb component become in-
creasingly prominent with the continued rise of the Rocky
Mountains and with the next major temperature decline
beginning in the Middle Miocene. An important factor in
the development and maintenance of savanna is reduced
and strongly seasonal rainfall, and the dry season occurs in
Table 6.18. Summary of North American vegetation types and estimated MATs, Middle Eocene through Early Miocene.
Region Vegetation MAT

Middle Eocene
Southeast (-30° N)
Claiborne, Jackson
Coastal Tropical
Lowland Tropical dry, sand pine scrub, flood plain, deciduous 24°C (20- 28°C)
[elements of mixed mesophytic, southern mixed hardwood,
beech (Claiborne)-maple (Acer'?, Jackson), maple (Acer1?,
Jackson)-basswood (Tilia, Claiborne), oak-hickory,
oak-chestnut]
Upland Deciduous
Highland Early Appalachian Occasional frost
montane coniferous (Picea, Tsuga)
Northeast
Ellesmere Island (81° N) Mixed hard wood-coniferous (lake states type) 12-13°C
West
High Plains,
eastern Rocky Mts.
Little Mountain Savanna, woodland
Tipperary, Subtropical
Kisinger lakes
Green River
300m Savanna, woodland (oak, pinon), tropical dry forest 16°C
1300m Deciduous
1700m Mixed hardwood-coniferous
Volcanic highlands
Germer (600 m) Mixed hardwood-coniferous
Thunder Mt. (1550m) Mixed hardwood-coniferous to western montane coniferous
S. California
Elsinore
Lowland Paratropical, deciduous
Upland Mixed hard wood-coniferous
N. California
Susanville Tropical rain forest 27°
Northwest
Oregon
Clarno Tropical to paratropical CLAMP, 16°C
Washington
Republic Mixed hardwood-coniferous, early western montane coniferous 12-13°C
British Columbia
Princeton Mixed hardwood-coniferous
Alaska
Lowland Notophyllous broad-leaved evergreen CLAMP, 15-16°C

Inland, upland Deciduous, early mixed hardwood-coniferous

Late Eocene
West
High Plains, eastern
Rocky Mts.
Colorado
Florissant
1200m Savanna-woodland (oak, pine), tropical dry 18°C (CLAMP, 12.5°C)
Montana (Late Eocene
to earliest Oligocene)
Beaverhead Basins, Metzel
Ranch, Mormon Creek,
York, Ruby
Lowland Chaparral, woodland, deciduous
Upland Mixed hardwood-coniferous
Western montane coniferous
continued
Table 6.18. continued
Region Vegetation MAT

Volcanic highlands
Nevada
Copper Basin, Mixed hardwood-coniferous, 11°C
Bull Run western montane coniferous
(1200-1500 m)
N. California
La Porte, Lower Tropical rain? 24°C
Cedarville (CLAMP 20°C)
Northwest
Oregon
Comstock, Goshen Tropical rain 24°C (CLAMP 20°C)
Alaska
Coastal Subtropical, notophyllous broad-leaved evergreen? 16°C (CLAMP)
Inland, upland Deciduous, mixed hardwood-coniferous 15°C (CLAMP)

Oligocene
Southeast
Catahoula
Vicksburg
Coastal Tropical
Inland Tropical toward deciduous (mixed mesophytic)
Upland Deciduous (mixed mesophytic)
West
High Plains, eastern Shrubland, woodland, savanna
Rocky Mts.
Colorado
Creede
Lowland Chaparral (mountain mahogany), woodland (pinon-juniper)
Upland Western montane coniferous (fir-pine, fir-spruce) 4.5°C
Northwest
Oregon
Bridge Creek, Fossil Deciduous 10-12°C
Rujada
Coastal Notophyllous broad-leaved evergreen
Inland Deciduous
Upland Western montane coniferous
Alaska Deciduous 4.5-7°C
MacKenzie Delta
Richards Deciduous

Early Miocene
Southeast Warm-temperate to subtropical?
Northeast
Vermont
Brandon Deciduous (mixed mesophytic) 17°C
Devon Island
Haughton Mixed hardwood-coniferous (lake states type) 0-11°C
Banks Island
Mary Sachs Gravel Mixed hardwood-coniferous (lake states type) 8-12°C
West
High Plains Grassy savanna, woodland?
Northern Rocky Mts.
Lowland Deciduous
Upland Western montane coniferous?
Nevada
Buffalo Canyon
Lowland Woodland, shrubland, notophyllous broad-leaved evergreen 10°C
Upland Mixed hardwood-coniferous
Sutro
Lowland Notophyllous broad-leaved evergreen
Upland Western montane coniferous

Table 6.18. continued
Region Vegetation
S. California
Tehachapi
Lowland Shrubland
Upland Woodland (evergreen oak, cypress, pifion)
Oregon
Alvord Creek Mixed hardwood-coniferous
Northwest
Alaska
Seldovia Point Deciduous, mixed hard wood-coniferous
Most plant formation and association names follow terminology used for modern communities discussed in Chapter 1 and listed in Table 1.1.
Remnants of older communities are named according to the terminology discussed in Chapter 5 (Vegetation).
the cooler half of the year (Buffalo Canyon and related flo-
ras). Savanna vegetation has mostly disappeared from
North America through continued development of the
Grassland Formation and more recently by spread of adja-
cent woodlands. Fires that normally constrained woodland
expansion are now mostly prevented. The greatest extent
of savanna was in recent times as a transition community
between the grassland and the oak-hickory association of
the deciduous forest formation (Chapter 2).
By the Middle Eocene, scattered highlands in the Rocky
Mountains supported Picea, Pinus, and Betula. In the
moister volcanic highlands to the west Abies, Larix, Picea,
Thuja, Tsuga, Alnus, Castanea, Populus, Quercus, and
others were already established as a broad-leaved decidu-
ous forest, a mixed hardwood-conifer forest with many
present-day Asian representatives, and a western montane
coniferous forest at the highest elevations (above 1200 m in
Nevada-Idaho). The Republic flora preserves the oldest
record of co-occurring coniferous forest elements (Abies,
Picea, Pinus, Pseudolarix, Tsuga, Chamaecyparis, Thuja,
and Betula of the B. papyrifera-B. occidentalis complex).
Thus, the earliest presence of the Rocky Mountain Mon-
tane Coniferous Forest, like the less extensive and well-
defined Appalachian counterpart, can be placed in the
Middle Eocene. Many of these elements extended into the
Arctic at progressively lower elevations down to -300 m.
The mixed hardwood—coniferous forest is an ecotonal
community between the lower elevation deciduous forest
and the higher elevation montane coniferous forest. It is
more restricted in distribution today because of higher ele-
vations, sharper climatic gradients, and greater community
definition; however, in the Middle Eocene through the
Early Miocene it was widespread as a transitional vegeta-
tion type. During the Middle Eocene and Early Oligocene
microthermal angiosperm families such as the Aceraceae,
Betulaceae, and Rosaceae underwent rapid diversification
and throughout North America modernization of the vege-
tation to the generic level was well underway. In the cen-
tral Rocky Mountains, subtropical elements disappeared
during the Oligocene and the Cordilleran flora took on a
more modern aspect.
Middle Eocene through Early Miocene History 233
MAT
6-7°C
To the west from about central California northward to
just beyond the present Canadian border, more tropical
vegetation with Old World affinities fringed the coast.
During the drying trend of the Middle Eocene, early
preadapted components of a dry vegetation type appeared
(e.g., Ephedra, Celtis, Ocotea in the Green River flora).
These dry-habitat plants formed the shmbland/chapar-
ral-woodland-savanna formation in the western United
States during the Middle to Late Eocene, individual ele-
ments of which eventually diversified, coalesced, and ex-
panded into the desert formation with the rise of the Coast
Ranges, Cascade Mountains, and Sierra Nevada in the Neo-
gene.
Thus, by the end of the Early Miocene the older tropical
dry forest and much of the notophyllous broad-leaved
evergreen vegetation had disappeared. The principal plant
communities of North America were the remaining ele-
ments of a tropical community along the southern coasts,
deciduous forest (sand pine scrub and other components of
the pine woods association; oak-chestnut; oak-hickory;
southern mixed hardwoods; flood-plain forest), elements
of an Appalachian montane coniferous forest, lake states
forest (to the far north), shrubland/chaparral-wood-
land—savanna, mixed hardwood—conifer forest, and west-
ern montane coniferous forest. There is no evidence as yet
for extensive tundra, grassland, or desert; elements of the
boreal forest were only just being assembled in the western
volcanic highlands and in the high latitudes, preadapted
for the next downshift in temperatures that would begin in
the Middle Miocene.
References
Abbott, R L. 1981. Cenozoic paleosols, San Diego, Califor-
nia. Catena 8: 223-237.
Asaro, E et al. (+ 6 authors). 1988. Possible world-wide
Middle Miocene iridium anomaly and its relationship
to periodicity of impacts and extinctions. Global Cata-
strophes in Earth History: An Interdisciplinary Confer-
ence on Impacts, Volcanism, and Mass Mortality.
Lunar Planetary Institute, Houston, Tx. pp. 6-7.
234 Late Cretaceous and Cenozoic History of North American Vegetation
Axelrod, D. I. 1939. A Miocene flora from the western bor-
der of the Mohave Desert. Carnegie Inst. Wash. Con-
trib. Paleontol. 516:1-128.
. 1944. The Alvord Creek flora. Carnegie Inst. Wash.
Contrib. Paleontol. 553: 225-262.
. 1952. A theory of angiosperm evolution. Evolution
6: 29-60.
. 1957. Age-curve analysis of angiosperm floras. J.
Paleontol. 31: 273-280.
. 1961. How old are the angiosperms? Am. J. Sci.
259: 447-459.
. 1966a. A method for determining the altitudes of
Tertiary floras. Palaeobotanist 14: 144-171.
. 1966b. The Eocene Copper Basin flora of northeast-
ern Nevada. Univ. Calif. Publ. Geol. Sci. 59:1-125.
. 1966c. Potassium-argon ages of some western Ter-
tiary floras. Am. J. Sci. 264: 497-506.
. 1979. Age and origin of Sonoran Desert vegetation.
California Academy of Sciences, San Francisco. Occa-
sional Papers 132.
. 1984. An interpretation of Cretaceous and Tertiary
biota in polar regions. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 45: 105-147.
. 1985a. Rise of the grassland biome, central North
America. Bot. Rev. 51: 163-201.
. 1985b. Miocene floras from the Middlegate Basin,
west-central Nevada. Univ. Calif. Publ. Geol. Sci. 129.
. 1987. The Late Oligocene Creede flora, Colorado. Bennett, K. D. 1990. Milankovitch cycles and their effects
Univ. Calif. Publ. Geol. Sci. 130:1-235.
. 1990. Environment of the Middle Eocene (45 Ma)
Thunder Mountain flora, central Idaho. Natl. Geogr. Berggren, W. A., D. V. Kent, M.-R Aubry, and J. Hardenbol
Res. 6: 355-361.
. 1991. The Early Miocene Buffalo Canyon flora of
western Nevada. Univ. Calif. Publ. Geol. Sci. 135:
1-76.
. 1992. Climatic pulses, a major factor in legume
evolution. In: P. S. Herendeen and D. L. Dilcher (eds.),
Advances in Legume Systematics: Part 4. The Fossil
Record. Royal Botanic Gardens, Kew, U.K. pp.
259-279.
, M. T. K. Arroyo, and P. H. Raven. 1991. Historical
development of temperate vegetation in the Americas.
Rev. Chil. Hist. Nat. 64: 413-446.
Axelrod, D. I. and H. P. Bailey. 1976. Tertiary vegetation, cli-
mate, and altitude of the Rio Grande depression, New
Mexico-Colorado. Paleobiology 2: 235-254.
Barghoorn, E. S. 1950. The Brandon lignite. Vermont Bot. Brown, R. W. 1939. Fossil leaves, fruits, and seeds of Cer-
Bird Clubs Joint Bull. 18: 21-36.
. 1951. Age and environment: a survey of North
American Tertiary floras in relation to paleoecology. J.
Paleontol. 25: 736-744.
. 1953. Evidence of climatic change in the geologic
record of plant life. In: H. Shapley (ed.), Climatic
Change—Evidence, Causes, and Effects. Harvard Uni-
versity Press, Cambridge, MA. pp. 235-248.
Barron, E. J. 1983. A warm, equable Cretaceous: the nature Burnham, R. J. 1994. Paleoecological and floristic hetero-
of the problem. Earth-Sci. Rev. 19: 305-338.
Basinger, J. E 1976. Paleorosa similkameenensis, gen. et sp.
nov., permineralized flowers (Rosaceae) from the
Eocene of British Columbia. Can. J. Bot. 54: 2293- Buzek, C., Z. Kvacek, and S. R. Manchester. 1989. Sapinda-
2305.
. 1981. The vegetative body of Metasequoia milleri
from the Middle Eocene of southern British Columbia.
Can. J. Bot. 59: 2379-2410.
—. 1984. Seed cones of Metasequoia milleri from the
Middle Eocene of British Columbia. Can. J. Bot. 62:
281-289.
—, D. R. Greenwood, and T. Sweda. 1994. Early Ter-
tiary vegetation of Arctic Canada and its relevance to
paleoclimatic interpretations. In: M. C. Boulter and
H. C. Fisher (eds.), Cenozoic Plants and Climates of the
Arctic. NATO ASI Series l:Global Environmental
Change. Springer-Verlag, Berlin. Vol. 27, pp. 175-198.
and G. W. Rothwell. 1977. Anatomically preserved
plants from the Middle Eocene (Allenby Formation) of
British Columbia. Can. J. Bot. 55: 1984-1990.
Becker, H. E 1960. The Tertiary Mormon Creek flora from
the upper Ruby River Basin in southwestern Montana.
Palaeontograph., Abt. B 107: 83-126.
. 1961. Oligocene Plants from the Upper Ruby River
Basin, Southwestern Montana. Geological Society of
America, Boulder, CO. Mem. 82.
. 1969. Fossil plants of the Tertiary Beaverhead
Basins in southwestern Montana. Palaeontograph.,
Abt. B 127: 1-142.
. 1972. The Metzel Ranch flora of the Upper Ruby
River Basin, southwestern Montana. Palaeontograph.,
Abt. B 141:1-61.
-. 1973. The York Ranch flora of the upper Ruby River
Basin, southwestern Montana. Palaeontograph. Abt. B,
143: 1-93.
on species in ecological and evolutionary time. Paleo-
biology 16:11-21.
(eds.). 1995. Geochronology, Time Scales and Global
Stratigraphic Correlation. Society of Sedimentary Ge-
ology, Tulsa, OK. Special Publ. 54.
and D. R. Prothero. 1992. Eocene-Oligocene cli-
matic and biotic evolution: an overview. In: D. R.
Prothero and W. A. Berggren (eds.), Eocene-Oligocene
Climatic and Biotic Evolution. Princeton University
Press, Princeton, N.J. pp. 1-28.
Berry, E. W. 1916. The Lower Eocene Floras of Southeast-
ern North America. U.S. Geological Survey, Reston,
VA. Prof. Paper 91, pp. 1-481.
Birkenmajer, K. 1987. Tertiary glacial and interglacial de-
posits, South Shetland Islands, Antarctica:
geochronology versus biostratigraphy (a progress re-
port). Bull. Pol. Acad. Sci. Earth Sci. 36: 133-145.
cidiphyllum. J. Paleontol. 13: 485-499.
. 1940. New species and changes of name in some
American fossil floras. J. Wash. Acad. Sci. 30: 344-
356.
. 1946. Alterations in some fossil and living floras. J.
Wash. Acad. Sci. 36: 344-355.
-. 1959. A bat and some plants from the Upper
Oligocene of Oregon. J. Paleontol. 33: 125-129.
geneity in the plant-fossil record—an analysis based
on the Eocene of Washington. Bull. U.S. Geol. Surv.
2085-B: 1-36.
ceous affinities of the Pteleaecarpum fruits from the
Tertiary of Eurasia and North America. Bot. Gaz. 150:
477-489.
Call, V. B. and D. L. Dilcher. 1992. Investigations of an-

giosperms from the Eocene of southeastern North
America: samaras of Fraxinus wilcoxiana Berry. Rev.
Palaeobot. Palynol. 74: 249-266.
—. 1994. Parvileguminophyllum coloradensis, a new
combination for Mimosites coloradensis Knowlton,
Green River Formation of Utah and Colorado. Rev.
Palaeobot. Palynol. 80: 305-310.
-. 1997. The fossil record of Eucommia (Eucommi-
aceae) in North America. Am. J. Bot. 84:798-814.
Call, V. B., D. M. Erwin, and R. A. Stockey. 1993. Further ob-
servations on Paleorosa similkameenensis (Rosaceae)
from the Middle Eocene Princeton Chert of British Co-
lumbia, Canada. Rev. Palaeobot. Palynol. 78: 277-291.
Call, V. B. and R. A. Stockey. 1988a. Permineralized fruits
and seeds from the Princeton Chert (Middle Eocene) of
British Columbia: Araceae. Am. J. Bot. 75:1099-1113.
. 1988b. Permineralized fruits and seeds from the
Princeton Chert (Middle Eocene) of British Columbia:
Lythraceae. Can. J. Bot. 66: 303-312.
. 1989. Permineralized fruits and seeds from the
Princeton Chert (Middle Eocene) of British Columbia:
Nymphaeaceae. Bot. Gaz. 150: 207-217.
. 1990a. Permineralized fruits and seeds from the
Princeton Chert (Middle Eocene) of British Columbia:
Vitaceae. Can. J. Bot. 68: 288-295.
. 1990b. Vegetative remains of the Magnoliaceae
from the Princeton Chert (Middle Eocene) of British
Columbia. Can. J. Bot. 68:1327-1339.
. 1990c. Vegetative remains of the Rosaceae from the
Princeton Chert (Middle Eocene) of British Columbia.
Int. Assoc. Wood Anat. Bull. 11: 261-280.
. 1991. Fruits and seeds from the Princeton Chert
(Middle Eocene) of British Columbia: Rosaceae
(Prunoideae). Bot. Gaz. 152: 369-379.
and K. B. Pigg. 1991. The Princeton Chert: evidence
for in situ aquatic plants. Rev. Palaeobot. Palynol. 70:
173-185.
Cevallos-Ferriz, S. R. 1995. Fruits ofRibes from the Prince-
ton Chert, British Columbia, Canada. Am. J. Bot. 82: 84.
Chaney, R. W. 1924. A comparative study of the Bridge
Creek flora and the modern redwood forest. Carnegie
Inst. Wash. Contrib. Palaeontol. 349: 1-22.
. 1927. Geology and palaeontology of the Crooked
River Basin, with special reference to the Bridge Creek
flora. Carnegie Inst. Wash. Contrib. Palaeontol. 346:
45-138.
and E. I. Sanborn. 1933. The Goshen flora of west
central Oregon. Carnegie Inst. Wash. Contrib. Paleon-
tol. 439.
Christe, R. L. 1964. Geological Reconnaissance of North-
eastern Ellesmere Island, District of Franklin, Canada.
Geological Survey of Canada, Ottawa, Memoir 331.
and G. E. Rouse. 1976. Eocene Beds at Lake Hazen,
Northern Ellesmere Island. Geological Survey of
Canada, Ottawa, Report on Activities Paper 76-3C,
pp. 153-156.
Collinson, M. E., K. Fowler, and M. C. Boulter. 1981. Floris-
tic changes indicate a cooling climate in the Eocene of
southern England. Nature 291: 315-317.
Cope, M. J. 1981. Products of natural burning as a compo-
nent of the dispersed organic matter of sedimentary
rocks. In: J. Brooks (ed.), Organic Maturation Studies
and Fossil Fuel Exploration. Academic Press, New
York. pp. 89-109.
Middle Eocene through Early Miocene History 235
Crane, R R. and R. A. Stockey. 1985. Growth and reproduc-
tive biology of foffrea speirsii gen. et sp. nov., a Cer-
cidiphyllum-li]ae plant from the Late Paleocene of Al-
berta, Canada. Can. J. Bot. 63: 340-364.
. 1987. Betula leaves and reproductive structures
from the Middle Eocene of British Columbia, Canada.
Can. J. Bot. 65: 2490-2500.
Crepet, W. L. and C. R Daghlian. 1980. Castaneoid inflores-
cences from the Middle Eocene of Tennessee and the
diagnostic value of pollen (at the subfamily level) in
the Fagaceae. Am. J. Bot. 67: 739-757.
. 1982. Euphorbioid inflorescences from the Middle
Eocene Claiborne Formation. Am. J. Bot. 69: 258-266.
and M. Zavada. 1980. Investigations of angiosperms
from the Eocene of North America: a new juglanda-
ceous catkin. Rev. Palaeobot. Palynol. 30: 361-370.
Crepet, W. L. and D. L. Dilcher. 1977. Investigations of an-
giosperms from the Eocene of North America: a mi-
mosoid inflorescence. Am. J. Bot. 64: 714-725.
Crepet, W. L., P. S. Herendeen, and E W. Potter, Jr. 1975. In-
vestigations of angiosperms from the Eocene of North
America: a catkin with juglandaceous affinities. Am. J.
Bot. 62: 813-823.
Crepet, W. L. and T. E Stuessy. 1978. A reinvestigation of
the fossil Viguiera cronquistii (Compositae). Brittonia
30:483-491.
Daghlian, C. P. and W. L. Crepet. 1983. Oak catkins, leaves
and fruits from the Oligocene Catahoula Formation and
their evolutionary significance. Am. J. Bot. 70:
639-649.
Crepet, W. L. and T. Delevoryas. 1980. Investigations of Ter-
tiary angiosperms: a new flora including Eomi-
mosoidea plumosa from the Oligocene of eastern
Texas. Am. J. Bot. 67: 309-320.
Crepet, W. L. and D. L. Dilcher. 1972. Middle Eocene sa-
baloid palms. Indiana Acad. Sci. Proc. 81: 94.
Dickinson, K. A. and J. S. Leventhal. 1988. The geology,
carbonaceous materials, and origin of the epigenetic
uranium deposits in the Tertiary Sespe Formation in
Ventura County, California. Bull. U.S. Geol. Surv. 1771:
1-18.
Dilcher, D. L. 1963. Cuticular analysis of Eocene leaves of
Ocotea obtusifolia. Am. J. Bot. 50:1-8.
. 1968. Revision of Eocene palms from southeastern
North America based upon cuticular analysis. Am. J.
Bot. 55: 725.
. 1969. Podocarpus from the Eocene of North Amer-
ica. Science 164: 299-301.
. 1973. A paleoclimatic interpretation of the Eocene
floras of southeastern North America. In: A. Graham
(ed.),Vegetation and Vegetational History of Northern
Latin America. Elsevier Science Publishers, Amster-
dam, pp. 39-59.
and C. R Daghlian. 1977. Investigations of an-
giosperms from the Eocene of southeastern North
America: Philodendron leaf remains. Am. J. Bot. 64:
526-534.
Dilcher, D. L. and G. E. Dolph. 1970. Fossil leaves of Den-
dropanax from Eocene sediments of southeastern
North America. Am. J. Bot. 57:153-160.
Dilcher, D. L. and S. R. Manchester. 1986. Investigations of
angiosperms from the Eocene of North America: leaves
of the Engelhardieae (Juglandaceae). Bot. Gaz. 147:
189-199.
236 Late Cretaceous and Cenozoic History of North American Vegetation
. 1988. Investigations of angiosperms from the
Eocene of North America: a fruit belonging to the Eu-
phorbiaceae. Tert. Res. 9: 45-58.
Dilcher, D. L. and J. E McQuade. 1967. A morphological
study of Nyssa endocarps from Eocene deposits in
western Tennessee. Bull. Torrey Bot. Club 94: 35-40.
Dilcher, D. L., E W. Potter, Jr., and W. L. Crepet. 1976. Inves-
tigations of angiosperms from the Eocene of North
America: juglandaceous winged fruits. Am. J. Bot. 63:
532-544.
Epis, R. C. and C. E. Chapin. 1974. Stratigraphic nomencla-
ture of the Thirtynine Mile volcanic field, central Col-
orado. Bulletin U.S. Geological Survey 1395-C: 1-22.
Erwin, D. M. 1987. Silicified palm remains from the Mid-
dle Eocene (Allenby Formation) of British Columbia.
Am. J. Bot. 74: 681.
and R. A. Stockey. 1989. Permineralized mono-
cotyledons from the Middle Eocene Princeton Chert
(Allenby Formation) of British Columbia: Alismat-
aceae. Can. J. Bot. 67: 2636-2645.
. 1990. Sapindaceous flowers from the Middle
Eocene Princeton Chert (Allenby Formation) of British
Columbia, Canada. Can. J. Bot. 68: 2025-2034.
. 1991a. Soleredera rhizomorpha gen. et sp. nov., a
permineralized monocotyledon from the Middle
Eocene Princeton Chert of British Columbia, Canada.
Bot. Gaz. 152: 231-247.
. 199lb. Silicified monocotyledons from the Middle
Eocene Princeton Chert (Allenby Formation) of British
Columbia, Canada. Rev. Palaeobot. Palynol. 70: 147-
162.
. 1992. Vegetative body of a permineralized mono-
cotyledon from the Middle Eocene Princeton Chert of
British Columbia. Cour. Forsch.-Inst. Senckenberg 147:
309-327.
-. 1994. Permineralized monocotyledons from the
Middle Eocene Princeton Chert (Allenby Formation) of
British Columbia, Canada: Arecaceae. Paleontograph.,
Abt. B 234: 19-40.
Estes, R. and J. H. Hutchison. 1980. Eocene lower verte-
brates from Ellesmere Island, Canadian Arctic Archi-
pelago. Palaeogeogr., Palaeoclimatol., Palaeoecol. 30:
325-347.
Eugster, H. R 1982. Climatic significance of lake and evap-
orite deposits. In: Geophysics Study Committee (eds.),
Climate in Earth History. National Academy Press,
Washington, B.C. pp. 105-111.
and L. A. Hardie. 1978. Saline lakes. In: A. Lerman
(ed.), Lakes—Chemistry, Geology, Physics. Springer-
Verlag, Berlin, pp. 237-293.
Evanoff, E., D. R. Prothero, and R. H. Lander. 1992.
Eocene-Oligocene climatic change in North America:
the White River Formation near Douglas, east-central
Wyoming. In: D. R. Prothero and W. A. Berggren (eds.),
Eocene-Oligocene Climatic and Biotic Evolution.
Princeton University Press, Princeton, N.J. pp. 116-
130.
Evernden, J. E, D. E. Savage, G. H. Curtis, and G. T. James.
1964. Potassium-argon dates and the Cenozoic mam-
malian chronology of North America. Am. J. Sci. 262:
145-198.
Eyde, R. H. 1997. Fossil record and ecology of Nyssa (Cor-
naceae), Bot. Rev. 63: 97-123.
Eyde, R. H. and E. S. Barghoorn. 1963. Morphological and
paleobotanical studies of the Nyssaceae, II the fossil
record. J. Arnold Arb. 44: 328-370.
Eyde, R. H., A. Bartlett, and E. S. Barghoorn. 1969. Fossil
record of Alangium. Bull. Torrey Bot. Club 96: 288-
314.
Folk, R. L. 1955. Petrographic Stratigraphy of the Texas
Gulf Coast Tertiary. Guidebook, Tertiary Field Trip.
University of Texas Geological Society, Austin, TX. pp.
15-18.
Forest, C. E., P Molnar, and K. A. Emanuel. 1995. Palaeoal-
timetry from energy conservation principles. Nature
374: 347-350.
Frederiksen, N. O. 1980a. Sporomorphs from the Jackson
Group (Upper Eocene) and Adjacent Strata of Missis-
sippi and Western Alabama. U.S. Geological Survey,
Reston, VA. Professional Paper 1084, pp. 1-75.
. 1980b. Mid-Tertiary climate of southeastern United
States: the sporomorph evidence. J. Paleontol. 54:
728-739.
. 1981. Middle Eocene to Early Oligocene plant com-
munities of the Gulf Coast. In: J. Gray, A. J. Boucot, and
W. B. Berry (eds.), Communities of the Past. Hutchin-
son Ross Publ. Co., Stroudsburg, PA. pp. 493-549.
. 1988. Sporomorph Biostratigraphy, Floral Changes,
and Paleoclimatology, Eocene and Earliest Oligocene
of the Eastern Gulf Coast. U.S. Geological Survey,
Reston, VA. Professional Paper 1448, pp. 1-68.
. 1991. Pulses of Middle Eocene to earliest
Oligocene climatic deterioration in southern Califor-
nia and the Gulf Coast. Palaios 6: 564-571.
Gandolfo, M. 1996. The presence of Fagaceae (oak family)
in sediments of the Klondike Mountain Formation
(Middle Eocene), Republic, Washington. Wash. Geol.
24: 20-21.
Gee, C. T. 1990. On the fossil occurrences of the mangrove
palm Nypa. Proceedings of the Symposium, Paleo-
floristic and Paleoclimatic Changes in the Cretaceous
and Tertiary, Prague, 1989. pp. 315-319.
Geophysics Study Committee. 1982. Climate in Earth His-
tory. National Academy Press, Washington, D.C.
Graham, A. 1996. A contribution to the geologic history of
the compositae. In: D. J. N. Hind and H. J. Beentje
(eds.), Compositae. Royal Botanic Gardens, Kew. pp.
123-140.
Gray, J. 1960. Temperate pollen genera in the Eocene (Clai-
borne) flora, Alabama. Science 132: 808-810.
Gregory, K. M. 1994. Palaeoclimate and palaeoelevation of
the 35 Ma Florissant flora, Front Range, Colorado.
Palaeoclimates 1: 23-57.
. 1996. Paleoclimate estimates biased by foliar phys-
iognomic responses to increased atmospheric CO2?
Palaeogeogr., Palaeoclimatol., Palaeoecol. 124: 39-51.
and C. G. Chase. 1992. Tectonic significance of pa-
leobotanically estimated climate and altitude of the
Late Eocene erosion surface, Colorado. Geology 20:
581-585.
Gregory, K. M. and W. C. Mclntosh. 1996. Paleoclimate and
paleoelevation of the Oligocene Pitch-Pinnacle flora,
Sawatch Range, Colorado. Geol. Soc. Amer. Bull. 108:
545-561.
Gregory-Wodzicki, K. M. 1997. The Late Eocene House
Ranch Flora, Sevier Desert, Utah: Paleoclimate and Pa-
leoelevation. Palaios 12: 552-567.
Haq, B. U, J. Hardenbol, and R R. Vail. 1987. Geochronol-

ogy of fluctuating sea levels since the Triassic. Science
235: 1156-1166.
Herendeen, P S. and D. L. Dilcher. 1990a. Reproductive and
vegetative evidence for the occurrence of Crudia
(Leguminosae, Caesalpinioideae) in the Eocene of
southeastern North America. Dot. Gaz. 151: 402-413.
. 1990b. Fossil mimosoid legumes from the Eocene
and Oligocene of southeastern North America. Rev.
Palaeobot. Palynol. 62: 339-361.
. 1990c. Diplotropis (Leguminosae, Papilionoideae)
from the Middle Eocene of southeastern North Amer-
ica. Syst. Bot. 15: 526-533.
-. 1991. Caesalpinia subgenus Mezoneuron (Legumi-
nosae, Caesalpinioideae) from the Tertiary of North
America. Am. J. Bot. 78: 1-12.
Herendeen, P. S., D. H. Les, and D. L. Dilcher. 1990. Fossil
Ceratophyllum (Ceratophyllaceae) from the Tertiary of
North America. Am. J. Bot. 77: 7-16.
Hickey, L. J., K. R. Johnson, and M. R. Dawson. 1988. The
stratigraphy, sedimentology, and fossils of the
Haughton Formation: a post-impact crater fill, Devon
Island, N.W.T., Canada. Meteoritics 23: 221-231.
Hut, P. et al. (+ 7 authors). 1987. Comet showers as a cause
of mass extinctions. Nature 329: 118-126.
Jones, E. L. 1961. Plant microfossils of the Laminated Sed-
iments of the Lower Eocene Wilcox Group in South-
Central Arkansas. Thesis, University of Oklahoma,
Norman, OK.
Kennett, J. P. and L. D. Stott. 1991. Abrupt deep-sea warm-
ing, palaeoceanographic changes and benthic extinc-
tions at the end of the Paleocene. Nature 353: 225-
229.
Koch, E L., J. C. Zachos, and P. D. Gingerich. 1992. Correla-
tion between isotope records in marine and continen-
tal carbon reservoirs near the Palaeocene/Eocene
boundary. Nature 358: 319-322.
Koeberl, C., C. Poag, W. Reimold, and D. Brandt. 1996. Im-
pact origin of the Chesapeake Bay structure and the
source of the North American tektites. Science 271:
1263-1266.
Kvacek, Z., C. Buzek, and S. R. Manchester. 1991. Fossil
fruits of Pteleaecarpum Weyland—Tiliaceous, not
Sapindaceous. Bot. Gaz. 152: 522-523.
Lakhanpal, R. N. 1958. The Rujada flora of west central
Oregon. Univ. Calif. Publ. Geol. Sci. 35: 1-66.
Leopold, E. B., G. Liu, and S. Clay-Poole. 1992. Low-
biomass vegetation in the Oligocene? In: D. R. Prothero
and W. A. Berggren (eds.), Eocene-Oligocene Climatic
and Biotic Evolution. Princeton University Press,
Princeton, NJ. pp. 399-420.
and H. D. MacGinitie. 1972. Development and
affinities of Tertiary floras in the Rocky Mountains. In:
A. Graham (ed.), Floristics and Paleofloristics of Asia
and Eastern North America. Elsevier Science Publish-
ers, Amsterdam.pp. 147-200.
MacGinitie, H. D. 1941. A Middle Eocene flora from the
central Sierra Nevada. Carnegie Inst. Wash. Contrib.
Paleontol. 534.
. 1953. Fossil plants of the Florissant beds, Colorado.
Carnegie Inst. Wash. Contrib. Paleontol. 599.
. 1969. The Eocene Green River flora of northwest-
ern Colorado and northeastern Utah. Univ. Calif. Publ.
Geol. Sci. 83:1-203.
. 1974. An early Middle Eocene flora from the Yel-
Middle Eocene through Early Miocene History 237
lowstone-Absaroka volcanic province, northwestern
Wind River Basin, Wyoming. Univ. Calif. Publ. Geol.
Sci. 108: 1-103.
Magallon-Puebla, S. and S. R. S. Cevallos-Ferriz. 1994. Eu-
commia constans N. Sp. fruits from upper Cenozoic
strata of Puebla, Mexico: Morphological and anatomi-
cal comparison with Eucommia ulmoides Oliver. In-
ternat. J. Plant Sci. ISS: 80-95.
Manchester, S. R. 1986. Vegetative and reproductive mor-
phology of an extinct plane tree (Platanaceae) from the
Eocene of western North America. Bot. Gaz. 147:
200-226.
. 1989. Attached reproductive and vegetative re-
mains of the extinct American-European genus Ce-
drelospermum (Ulmaceae) from the Early Tertiary of
Utah and Colorado. Am. J. Bot. 76: 256-276.
. 1990. Eocene to Oligocene floristic changes
recorded in the Clarno and John Day Formations, Ore-
gon, USA. Proceedings of the Symposium, Paleofloris-
tic and Paleoclimatic Changes in the Cretaceous and
Tertiary, Prague, 1989. pp. 183-187.
. 1991. Cruciptera, a new juglandaceous winged
fruit from the Eocene and Oligocene of western North
America. Syst. Bot. 16: 715-725.
. 1992. Flowers, fruits, and pollen of Florissantia, an
extinct malvalean genus from the Eocene and
Oligocene of western North America. Am. J. Bot. 79:
996-1008.
. 1994. Fruits and seeds of the Middle Eocene Nut
Beds flora, Clarno Formation, Oregon. Palaeontograph.
Am. 58: 1-205.
-. In press. Update on the Megafossil Flora of Floris-
sant, Colorado, USA. Geological Society of America,
Boulder, CO. Special Paper.
Manchester, S. R., M. E. Collinson, and K. Goth. 1994.
Fruits of the Juglandaceae from the Eocene of Messel,
Germany, and implications for Early Tertiary phyto-
geographic exchange between Europe and western
North America. Int. J. Plant Sci. 155: 388-394.
Manchester, S. R. and P. R. Crane. 1983. Attached leaves,
inflorescences, and fruits o^Fagopsis, an extinct genus
of fagaceous affinity from the Oligocene Florissant
flora of Colorado, U.S.A. Am. J. Bot. 70:1147-1164.
. 1987. A new genus of Betulaceae from the
Oligocene of western North America. Bot. Gaz. 148:
263-273.
Manchester, S. R., D. L. Dilcher, and W. D. Tidwell. 1986.
Interconnected reproductive and vegetative remains of
Populus (Salicaceae) from the Middle Eocene Green
River Formation, northeastern Utah. Am. J. Bot. 73:
156-160.
Manchester, S. R. and M. J. Donoghue. 1995. Winged fruits
of Linnaceae (Caprifoliaceae) in the Tertiary of western
North America: Diplodipelta gen nov. Int. J. Plant Sci.
156: 709-722.
Manchester, S. R. and W. J. Kress. 1993. Fossil bananas
(Musaceae): Ensete oregonense sp. nov. from the
Eocene of western North America and its phytogeo-
graphic significance. Am. J. Bot. 80: 1264-1272.
Manchester, S. R. and H. W. Meyer. 1987. Oligocene fossil
plants of the John Day Formation, Fossil, Oregon. Ore-
gon Geol. 49: 115-127.
Manchester, S. R. and E. A. Wheeler. 1993. Extinct juglan-
daceous wood from the Eocene of Oregon and its im-
238 Late Cretaceous and Cenozoic History of North American Vegetation
plications for xylem evolution in the Juglandaceae.
Int. Assoc. Wood Anat. Bull. 14: 103-11.
Marty, R., R. Dunbar, J. B. Martin, and P. Baker. 1988. Late
Eocene diatomite from the Peruvian coastal desert,
coastal up welling in the eastern Pacific, and Pacific
circulation before the terminal Eocene event. Geology
16:818-822.
Matthews, J. V, Jr. 1987. Plant Macrofossils from the Neo-
gene Beaufort Formation on Banks and Meighen is-
lands, District of Franklin. Current Research, part A.
Geological Survey of Canada, Ottawa, Paper 87-1A,
pp. 73-87.
and L. E. Ovenden. 1990. Late Tertiary plant macro-
fossils from localities in Arctic/Subarctic North Amer-
ica: a review of the data. Arctic 43: 364-392.
McClammer, J. U. 1978. Paleobotany and Stratigraphy of
the Yaquina Flora (Latest Oligocene-Earliest Miocene)
of Western Oregon. Thesis, University of Maryland,
College Park, MD.
Mclntyre, D. J. 1991. Pollen and spore flora of an Eocene
forest, eastern Axel Heiberg Island, N.W.T. Geol. Surv.
Can. Bull. 403: 83-97.
McKenna, M. C. 1980. Eocene paleolatitude, climate, and
mammals of Ellesmere Island. Palaeogeogr., Palaeocli-
matol, Palaeoecol. 30: 349-362.
. 1983. Holarctic landmass rearrangement, cosmic
events, and Cenozoic terrestrial organisms. Ann. Mis-
souri Bot. Garden 70: 459-489.
McLeroy, C. A. and R. Y. Anderson. 1966. Laminations of
the Oligocene Florissant Lake deposits, Colorado.
Geol. Soc. Am. Bull. 77: 605-618.
Meyer, H. W. 1986. An Evaluation of Methods for Estimat-
ing Paleoaltitudes Using Tertiary Floras from the Rio
Grande Rift Vicinity, New Mexico and Colorado. Ph.D.
dissertation, University California, Berkeley, CA.
. 1992. Lapse rates and other variables applied to es-
timating paleoaltitudes from fossil floras. Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 99: 71-99.
and S. R. Manchester. 1996. Paleobotanical signifi-
cance of Eocene flowers, fruits, and seeds from Repub-
lic, Washington. Wash. Geol. 24: 25-27.
-. 1997. The Oligocene Bridge Creek Flora of the John
Day Formation, Oregon. Univ. Calif. Publs. Geol. Sci.
141:1-19.
Miller, C. N., Jr. 1973. Silicified cones and vegetative re-
mains of Pinus from the Eocene of British Columbia.
Contrib. Univ. Mich. Mus. Paleontol. 24: 101-118.
Miller, K. G. et al. (+ 6 authors). 1990. Eocene-Oligocene
sea-level changes on the New Jersey coastal plain
linked to the deep-sea record. Geol. Soc. Am. Bull.
102: 331-339.
Myers, J. A. 1990. A Bridgerian Age Flora from Del Mar,
California. Master's thesis, San Diego State University,
San Diego, CA.
. 1996. Volcanic arcs and vegetation. Wash. Geol. 24:
37-39.
Nichols, D. J. 1970. Palynology in relation to depositional
environments of lignite in the Wilcox Group (Early
Tertiary) in Texas. Thesis, Pennsylvania State Univer-
sity, University Park, PA.
Norris, G. 1982. Spore-pollen evidence for Early Oligocene
high-latitude cool climatic episode in northern
Canada. Nature 297: 388-389.
Novacek, M. J. and J. A. Lillegraven. 1979. Terrestrial verte-
brates from the later Eocene of San Diego County, Cal-
ifornia: a conspectus. In: E L. Abbott (ed.), Eocene De-
positional Systems, San Diego, California. Pacific Sec-
tion, Society for Economic Paleontology and
Mineralogy, Los Angeles, pp. 69-79.
Omar, G. et al. (+ 5 authors). 1987. Fission-track dating of
Haughton astrobleme and included biota, Devon Is-
land, Canada. Science 237:1603-1605.
Owens, J. P. et al. (+ 5 authors). 1988. Stratigraphy of the
Tertiary Sediments in a 945-Foot-Deep Corehole Near
Mays Landing in the Southeastern New Jersey Coastal
Plain. U.S. Geological Survey, Reston, VA. Professional
Paper 1484, pp. 1-39.
Parrish, J. T, A. M. Ziegler, and C. R. Scotese. 1982. Rainfall
patterns and the distribution of coals and evaporites in
the Mesozoic and Cenozoic. Palaeogeogr., Palaeoclima-
tol., Palaeoecol. 40: 67-101.
Peterson, G. L. and R L. Abbott. 1979. Mid-Eocene climatic
change, southwestern California and northwestern
Baja California. Palaeogeogr., Palaeoclimatol., Palaeo-
ecol. 26: 73-87.
Phipps, C. J., J. M. Osborn, and R. A. Stockey. 1995. Pinus
pollen cones from the Middle Eocene Princeton Chert
(Allenby Formation) of British Columbia, Canada. Int.
J. Plant Sci. 156: 117-124.
Pigg, K. B. and R. A. Stockey. 1996. The significance of the
Princeton Chert permineralized flora to the Middle
Eocene upland biota of the Okanogan Highlands.
Wash. Geol. 24: 32-36.
and S. L. Maxwell. 1993. Paleomyrtinaea, a new
genus of permineralized myrtaceous fruits and seeds
from the Eocene of British Columbia and Paleocene of
North Dakota. Can. J. Bot. 71:1-9.
Pomerol, C. (ed.). 1981. Paleogene paleogeography and the
geological events at the Eocene-Oligocene boundary.
Palaeogeogr., Palaeoclimatol., Palaeoecol. 36:155-362.
and I. Premoli-Silva (eds.). 1986. Terminal Eocene
Events. Elsevier Science Publishers, Amsterdam.
Potbury, S. S. 1935. The La Porte flora of Plumas County,
California. Carnegie Inst. Wash. Contrib. Paleontol.
465: 29-81.
Povey, D. A. R., R. A. Spicer, and P C. England. 1994.
Palaeobotanical investigation of Early Tertiary palaeo-
elevations in northeastern Nevada: initial results. Rev.
Palaeobot. Palynol. 81: 1-10.
Prothero, D. R. 1985. North American mammalian diver-
sity and Eocene-Oligocene extinctions. Paleobiology
11: 389-405.
. 1989. Stepwise extinctions and climatic decline
during the later Eocene and Oligocene. In: S. K. Dono-
van (ed.), Mass Extinctions: Processes and Evidence.
Columbia University Press, New York. pp. 217-234.
. 1994. The Late Eocene-Oligocene extinctions.
Annu. Rev. Earth Planet. Sci. 22: 145-165.
. 1995. Geochronology and magnetostratigraphy of
Paleogene North American Land Mammal "Ages": an
update. In: W. A. Berggren, D. V Kent, M.-R Aubry, and
J. Hardenbol (eds.), Geochronology, Time Scales and
Global Stratigraphic Correlation. Society for Sedimen-
tary Geology, Tulsa, OK. Special Publ. 54, pp. 305-
315.
and W. A. Berggren (eds.). 1992. Eocene-
Oligocene Climatic and Biotic Evolution. Princeton
University Press, Princeton, NJ.

Rea, D. K., J. C. Zachos, R. M. Owen, and P. D. Gingerich.
1990. Global change at the Paleocene-Eocene bound-
ary: climatic and evolutionary consequences of tec-
tonic events. Palaeogeogr., Palaeoclimatol., Palaeoecol.
79: 117-128.
Reid, E. M. 1920. A comparative review of Pliocene floras,
based on the study of fossil seeds. Quarterly Geol. Soc.
Lond. 76:145-161.
Retallack, G. J. 1983a. Late Eocene and Oligocene Paleosols
from Badlands National Park, South Dakota. Geological
Society of America, Boulder, CO. Special Paper 193.
. 1983b. A paleopedological approach to the inter-
pretation of terrestrial sedimentary rocks: the mid-
Tertiary fossil soils of Badlands National Park, South
Dakota. Geol. Soc. Am. Bull. 94: 823-840.
-. 1990. Soils of the Past: An Introduction to Paleope-
dology. Unwin-Hyman, Boston.
Robert, C. and H. Chamley. 1991. Development of Early
Eocene warm climates, as inferred from clay mineral
variations in oceanic sediments. Palaeogeogr., Palaeo-
climatol., Palaeoecol. 89: 315-331.
Robison, C. R. and C. P. Person. 1973. A silicified semi-
aquatic dicotyledon from the Eocene Allenby Forma-
tion of British Columbia. Can. J. Bot. 51: 1373-1377.
Roehler, H. W. 1992a. Introduction to Greater Green River
Basin Geology, Physiography, and History of Investiga-
tions. U.S. Geological Survey, Reston, VA. Professional
Paper 1506-A, pp. 1-14.
. 1992b. Correlation, Composition, Areal Distribu-
tion, and Thickness of Eocene Stratigraphic Units,
Greater Green River Basin, Wyoming, Utah, and Col-
orado. U.S. Geological Survey Professional Paper
1506-E, pp. 1-49.
1993. Eocene Climates, Depositional Environ-
ments, and Geography, Greater Green River Basin,
Wyoming, Utah, and Colorado. U.S. Geological Survey,
Reston, VA. Professional Paper 1506-F, pp. 1-74.
Rothwell, G. W. and J. E Basinger. 1979. Metasequoia mil-
leri n. sp., anatomically preserved pollen cones from
the Middle Eocene (Allenby Formation) of British Co-
lumbia. Can. J. Bot. 57: 958-970.
Rothwell, G. W, R. A. Stockey, and H. Nishida. 1994. Fili-
caleans of the Middle Eocene Princeton Chert: I. A
dryopterid species. Am. J. Bot. 81:101-102.
Sanborn, E. I. 1935. The Comstock flora of west central
Oregon. Carnegie Inst. Wash. Contrib. Paleontol. 465:
1-28.
Schorn, H. E. and W. C. Wehr. 1996. The conifer flora from
the Eocene uplands at Republic, Washington. Wash.
Geol. 24: 22-24.
Scott, R. A. 1954. Fossil fruits and seeds from the Eocene
Clarno Formation of Oregon. Palaeontograph., Abt. B
96:66-97.
, E. S. Barghoorn, and E. B. Leopold. 1960. How old
are the angiosperms? Am. J. Sci. 258A: 284-299.
Sloan, L. C. 1994. Equable climates during the Early
Eocene: significance of regional paleogeography for
North American climate. Geology 22: 881-884.
and E. J. Barren. 1990. "Equable" climates during
Earth history? Geology 18: 489-492.
-. 1992. A comparison of Eocene climate model results
to quantified paleoclimatic interpretations. Palaeo-
geogr., Palaeoclimatol., Palaeoecol. 93:183-202.
Spackman, W. 1949. The flora of the Brandon lignite: geo-
Middle Eocene through Early Miocene History 239
logical aspects and a comparison of the flora with its
modern equivalents. Dissertation, Harvard University,
Cambridge, MA.
Stockey, R. A. 1984. Middle Eocene Pinus remains from
British Columbia. Bot. Gaz. 145: 262-274.
. 1987. A permineralized flower from the Middle
Eocene of British Columbia. Am. J. Bot. 74: 1878-
1887.
and K. B. Pigg. 1991. Flowers and fruits ofPrinceto-
nia allenbyensis (Magnoliopsida; family indet.) from
the Middle Eocene Princeton Chert of British Colum-
bia. Rev. Palaeobot. Palynol. 70: 163-172.
-. 1994. Vegetative growth of Eorhiza arnoldii Robi-
son and Person from the Middle Eocene Princeton
Chert locality of British Columbia. Int. J. Plant Sci.
155:606-616.
Stucky, R. K. 1990. Evolution of land mammal diversity in
North America during the Cenozoic. Curr. Mammal. 2:
375-432.
Sun, Z. and R. A. Stockey. 1991. Lauraceous inflorescences
from the Middle Eocene Princeton Chert (Allenby For-
mation) of British Columbia, Canada. Am. J. Bot. 78:125.
Tanai, T. and J. A. Wolfe. 1977. Revisions of Ulmus and
Zelkova in the Middle and Late Tertiary of Western
North America. U.S. Geological Survey, Reston, VA.
Professional Paper 1026, pp. 1-14.
Thelin, G. P. and R. J. Pike. 1991. Landforms of the conter-
minous United States—a digital shaded-relief por-
trayal. U.S. Geological Survey, Reston, VA. Ser. Map
1-2206.
Thomasson, J. R. 1985. Miocene fossil grasses: possible
adaptation in reproductive bracts (lemma and palea).
Ann. Missouri Bot. Garden 72: 843-851.
Thompson, G. R., R. W. Fields, and D. Alt. 1982. Land-
based evidence for Tertiary climatic variations: north-
ern Rockies. Geology 10: 413-417.
Tiffney, B. H. 1977. Fruits and seeds of the Brandon lignite:
Magnoliaceae. Bot. J. Linn. Soc. 75: 299-323.
. 1979. Fruits and seeds of the Brandon lignite III.
Turpinia (Staphyleaceae). Brittonia 31: 39-51.
. 1980. Fruits and seeds of the Brandon lignite, V Ru-
taceae. J. Arnold Arb. 61: 1-40.
. 1981. Fruits and seeds of the Brandon lignite, VI.
Microdiptera (Lythraceae). J. Arnold Arb. 62: 487-
516.
. 1993. Fruits and seeds of the Tertiary Brandon lig-
nite. VII. Sargentodoxa (Sargentodoxaceae). Am. J.
Bot. 80: 517-523.
. 1994. Re-evaluation of the age of the Brandon lig-
nite (Vermont, USA) based on plant megafossils. Rev.
Palaeobot. Palynol. 82: 299-315.
and E. S. Barghoorn. 1976. Fruits and seeds of the
Brandon lignite. I. Vitaceae. Rev. Palaeobot. Palynol.
22: 169-191.
. 1979. Flora of the Brandon lignite. IV. Illiciaceae.
Am. J. Bot. 66: 321-329.
Tiffney, B. H. and K. K. Haggard. 1996. Fruits of Mastix-
ioideae (Cornaceae) from the Paleogene of western
North America. Rev. Palaeobot. Palynol. 92: 29-54.
Tiffney, B. H. and A. Traverse. 1994. The Brandon lignite
(Vermont) is of Cenozoic, not Cretaceous, age! North-
east. Geol. 16: 215-220.
Ting, W. S. 1975. Eocene pollen assemblage from Del Mar,
southern California. Geosci. Man 11: 159.
240 Late Cretaceous and Cenozoic History of North American Vegetation
Tralau, H. 1964. The genus Nypa van Wurmb. Konigl.
Svenska Vetensk-Akad. Handl. 10: 1-29.
Traverse, A. 1955. Pollen Analysis of the Brandon Lignite Wodehouse, R. P. 1932. Tertiary pollen; I. Pollen of the liv-
of Vermont. U.S. Dept. Interior, Washington, B.C. Bu-
reau of Mines Kept. Invest. 5151.
. 1994. Palynofloral geochronology of the Brandon
lignite of Vermont, USA. Rev. Palaeobot. Palynol. 82:
265-297.
Upchurch, G. R., Jr. and J. A. Wolfe. 1987. Mid-Cretaceous
to Early Tertiary vegetation and climate: evidence from
fossil leaves and woods. In: E. M. Friis, W. G. Chaloner,
and P R. Crane (eds.), The Origins of Angiosperms and
Their Biological Consequences. Cambridge University
Press, Cambridge, U.K. pp. 75-105.
Van der Graaff, N. A. and R Baas. 1974. Wood anatomical
variation in relation to latitude and altitude. Blumea
22: 101-121.
Vrba, E. S., G. H. Denton, T. C. Partridge, and L. H. Burckle
(eds.), 1995. Paleoclimate and Evolution, with Empha-
sis on Human Evolution. Yale University Press, New
Haven, CT.
Washington Geology. 1996. 24.
Webb, S. D. 1985. Main pathways of mammalian diversifi-
cation in North America. In: E G. Stehli and S. D. Webb
(eds.), The Great American Biotic Interchange. Plenum,
New York. pp. 201-217.
. 1989. The fourth dimension in North American ter-
restrial mammal communities. In: D. W. Morris, Z.
Abramsky, B. J. Fox, and M. R. Willig (eds.), Patterns in
the Structure of Mammalian Communities. Texas Tech
University, Lubbock, TX. Special Publ. 28, pp. 181-
203.
Wehr, W. C. 1995. Early Tertiary flowers, fruits, and seeds
from Washington state and adjacent areas. Wash. Geol.
23: 3-16.
and D. Q. Hopkins. 1994. The Eocene orchards and
gardens of Republic, Washington. Wash. Geol. 22:
27-34.
Wehr, W. C. and S. R. Manchester. 1996. Paleobotanical sig-
nificance of Eocene flowers, fruits, and seeds from Re-
public, Washington. Wash. Geol. 24: 25-27.
Westgate, J. W. and C. T. Gee. 1990. Paleoecology of a mid-
dle Eocene mangrove biota (vertebrates, plants, and in-
vertebrates) from southwest Texas. Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 78:163-177.
Wheeler, E. A. and J. Landon. 1992. Late Eocene (Chadron-
ian) dicotyledonous woods from Nebraska: evolution-
ary and ecological significance. Rev. Palaeobot. Paly-
nol. 74: 267-282.
and C. A. LaPasha. 1994. Woods of the Vitaceae—
fossil and modern. Rev. Palaeobot. Palynol. 80:
175-207.
Whitlock, C. and M. R. Dawson. 1990. Pollen and verte-
brates of the Early Neogene Haughton Formation,
Devon Island, Arctic Canada. Arctic 43: 324-330.
Wing, S. L. 1987. Eocene and Oligocene floras and vegeta-
tion of the Rocky Mountains. Ann. Missouri Bot. Gar-
den. 74: 748-784.
. 1996. Paleocene-Eocene floral and climate change
in the northern Rocky Mountains, USA. International
Organization of Palaeobotanists Fifth Quadrennial
Conference, Santa Barbara, CA, June-July, 1996. p.
113.
and D. R. Greenwood. 1993. Fossils and fossil cli-
mate: the case for equable continental interiors in the
Eocene. Phil. Trans. R. Soc. Lond. (B) 341: 243-252.
ing representatives of the Green River flora. Bull. Tor-
rey Bot. Club 59: 313-340.
. 1933. Tertiary pollen; II. Pollen of the Green River
oil shales. Bull. Torrey Bot. Club 60: 479-524.
Wolfe, J. A. 1971. Tertiary climatic fluctuations and meth-
ods of analysis of Tertiary floras. Palaeogeogr., Palaeo-
climatol., Palaeoecol. 9: 27-57.
. 1973. Fossil forms of Amentiferae. Brittonia 25:
334-355.
. 1977. Paleogene floras from the Gulf of Alaska Re-
gion. U.S. Geological Survey, Reston, VA. Professional
Paper 997, pp. 1-108.
. 1978. A paleobotanical interpretation of Tertiary
climates in the Northern Hemisphere. Am. Sci. 66:
694-703.
. 1981a. A Chronologic Framework for Cenozoic
Megafossil Floras of Northwestern North America and
Its Relation to Marine Geochronology. Geological Soci-
ety of America, Boulder, CO. Special Paper 184, pp.
39-47.
. 1981b. Paleoclimatic significance of the Oligocene
and Neogene floras of the northwestern United States.
In: K. J. Niklas (ed.), Paleobotany, Paleoecology, and
Evolution. Praeger, New York. Vol. 2, pp. 79-101.
. 1985. Distribution of major vegetational types dur-
ing the Tertiary. In: E. T. Sundquist and W. S. Broecker
(eds.), The Carbon Cycle and Atmospheric CO2: Nat-
ural Variations Archean to Present. American Geo-
physical Union, Washington, D.C. Monograph 32, pp.
357-376.
. 1986. Tertiary floras and paleoclimates of the
Northern Hemisphere. Univ. Tennessee Dept. Geol.
Sci. Studies Geol. 15: 182-196.
. 1987. An overview of the origins of the modern
vegetation and flora of the northern Rocky Mountains.
Ann. Missouri Bot. Garden. 74: 785-803.
. 1989. North American Eocene vegetation and its
climatic implications. EOS (Trans. Am. Geophys.
Union) 70: 375.
. 1992a. Climatic, floristic, and vegetational changes
near the Eocene/Oligocene boundary in North Amer-
ica. In: D. R. Prothero and W. A. Berggren (eds.),
Eocene-Oligocene Climatic and Biotic Evolution.
Princeton University Press, Princeton, N.J. pp.
421-436.
. 1992b. An analysis of present-day terrestrial lapse
rates in the western conterminous United States and
their significance to paleoaltitudinal estimates. U.S.
Geol. Surv. Bull. 1964: 1-35.
. 1993. A method of obtaining climatic parameters
from leaf assemblages. U.S. Geol. Surv. Bull. 2040:
1-71.
. 1994a. Alaskan Paleogene climates as inferred from
the CLAMP database. In: M. C. Boulter and H. C.
Fisher (eds.), Cenozoic Plants and Climates of the Arc-
tic. NATO ASI Series 1: Global Environmental Change.
Springer-Verlag, Berlin. Vol. 27, pp. 223-234.
. 1994b. Tertiary climatic changes at middle lati-
tudes of western North America. Palaeogeogr., Palaeo-
climatol., Palaeoecol. 108: 195-205.
and E. S. Barghoorn. 1960. Generic change in Ter-

tiary floras in relation to age. Am. J. Sci. 258A: 388-
399.
Wolfe, J. A. and R. Z. Poore. 1982. Tertiary marine and non-
marine climatic trends. In: Geophysics Study Commit-
tee (eds.), Climate in Earth History. National Academy
Press, Washington, D.C. pp. 154-158.
Wolfe, J. A. and H. E. Schorn. 1989. Paleoecologic, paleo-
climatic, and evolutionary significance of the Oligo-
cene Creede flora, Colorado. Paleobiology 15: 180-
198.
. 1990. Taxonomic revision of the Spermatopsida of
the Oligocene Creede flora, southern Colorado. U.S.
Geol. Surv. Bull. 1923: 1-40.
Wolfe, J. A., H. E. Schorn, C. E. Forest, and P. Molnar. 1997.
Paleobotanical evidence for high altitudes in Nevada
during the Miocene. Science 276:1672-1675.
Wolfe, J. A. and T. Tanai. 1980. The Miocene Seldovia Point
flora from the Kenai Group, Alaska. U.S. Geological
Survey, Reston, VA. Professional Paper 1105, pp.
1-52.
. 1987. Systematics, phylogeny, and distribution of
Acer (maples) in the Cenozoic of western North Amer-
ica. J. Fac. Sci., Hokkaido Univ. 22: 1-246.
Wolfe, J. A. and W. Wehr. 1987. Middle Eocene dicotyledo-
nous plants from Republic, northeastern Washington.
U.S. Geol. Surv. Bull. 1597: 1-25.
Zachos, J. C., B. P. Flower, and H. Paul. 1997. Orbitally
Middle Eocene through Early Miocene History 241
paced climate oscillations across the Oligocene/
Miocene boundary. Nature 388: 567-570.
Zavada, M. S. and W. L. Crepet. 1981. Investigations of an-
giosperms from the Middle Eocene of North America:
flowers of the Celtidoideae. Am. J. Bot. 68: 924-933.
General Readings
Gregory, K, M. 1994. New prospects in old bubbles. Nature
372: 407-408.
Prothero, D. R. 1994. The Eocene-Oligocene Transition,
Paradise Lost. Columbia University Press, New York.
Retallack, G. J., E. A. Bestland, and T. J. Fremd. 1996. Re-
constructions of Eocene and Oligocene plants and an-
imals of central Oregon. Oregon Geol. 58: 51-69.
Valdes, P. 1994. Damping seasonal variations. Nature 372:
221.
Note
1. Calcareous nanoplankton is the floating aquatic
community comprising small organisms (~5um) with cal-
citic walls. The Coccolithoporidae (unicellular green
algae) are the principal component and leave remains
(nanofossils) in the form of wall platelets called coccoliths
and discoasters.
 SEVEN

Middle Miocene through Pliocene

North American Vegetational History
16.3-1.6 Ma

CONTEXT SUMMARY from North Atlantic deep sea sediments (Chamley, 1979).
At -4.8-4.9 Ma global cooling and a marine regression of
During the Middle Miocene through the Pliocene the Ap- -40-50 m combined to isolate the Mediterranean Basin
palachian Mountains underwent continued erosion and from the ocean and to concentrate large volumes of salt as
approached modern elevations (Figs. 7.1-7.3). The Rocky water evaporated. The biota was destroyed, giving rise to
Mountains had undergone uplift to half or more of their the term Messinian salinity crisis. The evaporation in-
present elevation during the Late Cretaceous to Middle volved -40 times the present water volume of the Medi-
Eocene Laramide Revolution; after a lull during the Middle terranean. The large amounts of salts stored in the basin
Eocene through the Early Miocene, there was increased reduced ocean salinity by -6%. This allowed northern
tectonic activity beginning -12 Ma and especially between hemisphere seawater to freeze at a higher temperature, cre-
7 and 4 Ma. Locally some highlands may have approached ating more extensive sea ice and providing positive feed-
or attained modern elevations. The increasingly high moun- back to the trend of lowering temperatures.
tains and plateaus of Asia and North America deflected the The cooling trend continued into the Early Pliocene,
major air streams southward, bringing colder polar air into then warmer than present marine waters were evident by
the middle latitudes of North America. An extensive Ant- Planktic Foraminiferal Zone N19 (Middle Pliocene, -4 Ma;
arctic ice sheet further cooled ocean waters and con- Cronin and Dowsett, 1991); that was the last time the Earth
tributed to the spread of seasonally dry climates. The elim- was significantly warmer than at present (by ~3.5°C; see
ination of most of the Asian exotics from the North collection of papers on Pliocene climates in Marine Geology,
American flora dates to the Late Miocene-Pliocene as a re- Vol. 27, 1996). This period culminating in mid-Pliocene
sult of a decline in summer rainfall. The Sierra Nevada at- warmth, especially evident at the mid- and high latitudes,
tained about two-thirds of their present elevation within correlates with an increase in CO2 concentration to near
the past 10 Ma. They were appreciably elevated at ~5 Ma, modern levels (-350 ppm; Crowley, 1991) and also in-
stood at -2100 m at 3 Ma, and have risen -950 m since 3 volved greater ocean heat transport (Raymo et al., 1996). At
Ma (Huber, 1981). The California Coast Ranges and Cas- -3.1 Ma the sea level was an estimated 35 ± 18 m higher
cade Mountains attained significant heights by 3 Ma, and along the Atlantic Coast than it is now (Dowsett and Cro-
there was a rapid rise of the Alaska Range at -6 Ma. nin, 1990), and transgressive events were the largest of the
Temperatures increased between -18 and 16 Ma (Fig. Late Neogene (Krantz, 1991). The shoreline extended fur-
3.1). In the absence of major plate reorganization and in- ther inland than at any time since the Eocene (Cronin,
tense volcanic activity and with increased erosion from 1991). The formation of the Bering Strait at -3.4 Ma corre-
continued replacement of the dense evergreen forest by de- lates with this rise. There was only seasonal ice in the Arc-
ciduous forest and shrubland (increasing albedo), atmo- tic Basin, likely occurring during the months of winter
spheric CO2 concentration decreased and a sharp lowering darkness. Estimates of winter temperatures north of 70° N
of temperature occurred in the Middle Miocene between (e.g., Baffin Island, Canadian Archipelago) are 20-22°C
15 and 10 Ma (Figs. 3.1, 7.3). Eolian dust deposits in- warmer than at present, and summer temperatures were
creased in the Late Cenozoic, suggesting greater aridity 6-8°C warmer (Zubakov and Borzenkova, 1990). One ef-
(Rea et al., 1985). This is supported by kaolinite records fect was wetter (pluvial) climates in the Middle Pliocene

242
 MIDDLE-LATE MIOCENE TIME SCALE

Figure 7.1. Middle-Late Miocene time scale. Reprinted from Berggren et al. (1995) with the permission of the Society for Sedimentary Geology.
 PLIOCENE-PLEISTOCENE TIME SCALE

Figure 7.2. Pliocene-Pleistocene time scale. Reprinted from Berggren et al. (1995) with the permission of the Soci-
ety for Sedimentary Geology.
 Middle Miocene through Pliocene History 245

compared to widespread Mio-Pliocene aridity. There is

debate as to the extent that this warming affected the East
Antarctic ice sheet (Barrett et al., 1992; Marchant et al.,
1993).
Although local glaciers dating to the Late Cretaceous
were possibly present intermittently at high elevations in
the paleo-Brooks Range of Alaska, ice cover in the Arctic
Ocean began in the Late Miocene (6-5 Ma) and intensified
in the Late Pliocene when ice volume reached about one-
half to two-thirds that of the Quaternary glaciations (Crow-
ley and North, 1991; Curry and Miller, 1989; 3.0-2.1 Ma,
Repenning and Brouwers, 1992). Some ice was present on
Greenland at -7 Ma (Larsen et al., 1994), and the Green-
land ice sheet may have formed by ~3.2 Ma (Leg 105 Ship-
board Scientific Party, 1986). There was cooling between
-3.0 and 2.1 Ma, and the eastern North American Lauren-
tide ice sheet began to form -2.6 Ma, correlating with the
abrupt appearance of ice-rafted debris in Arctic Ocean sed-
iments that was probably from mountain glaciers. The first
record of permafrost, a frigid Arctic Ocean, and the largest
Pliocene glaciation is at 2.4 Ma; between 2.4 and 2.1 Ma
perennial ice may have developed at the poles. There are
tillites older than 2.1 m.y. in the Puget Sound region of
Washington (reversed-magnetized tills of the lower Matu-
yama Chron), older than 1.75 m.y. in the Cascade Range,
and -2 m.y. in the Yellowstone region of Wyoming. Glacia-
tion intensified between 2.2 and 2.1 Ma when ice briefly
extended down the Mississippi River Valley as far south as
Iowa (type C till, Afton, Iowa). This cold period ended
abruptly at -2.1 Ma and fluctuating warmer conditions ex-
isted between -1.7 and 1.2 Ma. At -1.1 Ma cold periods be-
came more severe, the Arctic Ocean had periods of perma-
nent ice cover, and by 850 ka ice was again extending
down the Mississippi River Valley initiating the multiple
glaciations in the conterminous United States.
The growing ice sheets increased albedo, which served
as a positive feedback to cooling temperatures. The Mi-
lankovitch cycles are increasingly evident in the younger,
widespread, and better preserved sediments; climates
downshifted in the now familiar stepwise fashion (Dowsett
and Loubere, 1992). As noted by Krantz, "During the [L]ate
Pliocene, global climate settled into a phase of low-ampli-
tude glacial-interglacial fluctuations with a 41-ka period
forced by the obliquity cycle. This pattern persisted until
the [M]iddle Pleistocene transition to a dominant 100-ka
cycle beginning at -0.7 Ma." (1991, p. 169).
By the end of the Pliocene the Gulf of Mexico coastline
had retreated to within -120 km (80 m) inland of its pres-
ent position and extended from just west of Houston,
Texas, along a line from Baton Rouge, Louisiana, to Mobile,
Alabama, and along the west coast of Florida. Much of the
Figure 7.3. Principal fossil floras mentioned in the text Florida Peninsula below -27° N latitude was still in-
plotted with general reference to the paleotemperature undated. The Atlantic Ocean extended inland -45 km
record (lower curve) and sea level (upper curve) for the (30 m) up to Savannah, Georgia, then more narrowly inland
Late Cretaceous through the Early Eocene, h, see Fig. 3.1. through the mid-Atlantic States and New England. The
246 Late Cretaceous and Cenozoic History of North American Vegetation

Hadley and Rossby atmospheric circulation patterns and
the seasonal meandering of the polar fronts were near their
modern positions and subsequently varied with the Mi-
lankovitch cycles.
Ungulate faunas were prominent in North America,
doubling every 3-5 m.y. from the Hemingfordian to the
Clarendonian NALMA. This suggests a change from wood-
land savanna toward grassland savanna in parts of the con-
tinent. A major change in the composition of North Amer-
ican mammal communities took place in the Mio-
Pliocene when two-thirds of the new appearances were in-
troductions from Asia, including some microtid rodents, a
flying squirrel, a relative of the lesser panda, a brown bear,
elk, hyaenid, and rhinocerotid. Also making their first ap-
pearance near the Gauss-Matuyama magnetopolarity
boundary (-2.8 Ma) were new immigrants from South
America (sloth, armadillo, porcupine). By the end of the
Pliocene there was a decline in the rich savanna ungulate
fauna, which was gradually replaced by smaller herbi-
vores.
VEGETATION
The southeastern United States and the adjacent continen-
tal shelf includes fossiliferous sediments in a petroleum-
rich region that has been drilled extensively for oil and gas
(Fig. 7.4, Table 7.1). Yet little is known about the vegeta-
tional history or terrestrial paleoenvironments of the Gulf
Coast region after Vicksburgian (Early Oligocene) time. It is
probable that exotic elements continued to disappear, trop-
ical elements decreased, warm-temperate deciduous forest
expanded at low to middle elevations, and the Appa-
lachian montane coniferous forest developed further in the
highlands. The deciduous forest reached its maximum
southern extent during and after the Middle Miocene cool-
ing, and elements were established in eastern Mexico by
the Pliocene (Graham, 1976a,b). Pollen of the xerophyte
Ephedra is present in eastern coastal plain deposits
through the Middle Miocene (Frederiksen, 1984). If it was
actually growing there, it likely persisted in areas of phys-
iologically arid, deep sandy soil, even during times of in-
creasing precipitation in the southeast during the Middle
Tertiary, but was eliminated with the additional impact of
lowering temperatures. Modern pollen rain studies show
that Ephedra pollen can be transported long distances.
In the Pliocene, changes in the vegetation of southwest-
ern peninsular Florida are evident from pollen and spore
assemblages in the Tamiami Formation (Pinecrest Beds,
3.5-2 Ma; Table 7.2). Others can be inferred from coastal
ostracode and benthic foraminifera faunas (Willard et al.,
1993). In the interval between -4.5 and 3.5 Ma pollen of
Pinus, Betula, Carya, Fagus, Fraxinus, Juglans, Liquid-
ambar, and Ulmus indicate conditions cooler than at pres-
ent. At about 3.5 Ma climates warmed to the mid-Pliocene
maximum, slightly above to modern values, as shown by
an increase in Pinus and by the decline or disappearance of
deciduous hardwoods. The Pliocene warmth was due to
some increase in atmospheric CO2 and to an increase in
oceanic heat transport from the tropics. An appreciable in-
crease in CO2 would probably warm tropical and subtropi-
cal regions generally and allow the introduction of tropical
elements into subtropical Florida. This is not shown in the
pollen and spore assemblages, and according to Raymo
and Rau (1992, 1993) the increase in CO2 in the mid-
Pliocene was not appreciable. If the mechanism was oce-
anic heat transport, it is likely coastal sites would warm
but not the interior to the extent of supporting tropical veg-
etation. There is evidence for this mechanism in that the
Isthmus of Panama closed at about this time (3.5 Ma; Coates
et al., 1992; Graham, 1992), which strengthened the flow of
the Gulf Stream.
The reading of paleoclimates from coastal assemblages
is complex, however, because an increase in the pace of
ocean currents produces another effect that cools local cli-
mates. Upwelling brings cold bottom waters to the surface
and can create a local cool interval during an otherwise
warming cycle. This was the case in southeastern coastal
Mexico where upwelling contributed to the presence of a
cool-temperate element in the Middle Pliocene Paraje Solo
flora (Graham, 1976a,b). Similarly, in Florida a catastrophic
death assemblage of seabirds (extinct cormorant and oth-
ers) is preserved in the Pinecrest beds where red tides as-
sociated with increased upwellings probably caused toxic
poisoning (Emslie and Morgan, 1994). The implication is
that tropical holdovers from earlier times were periodically
eliminated and occasionally reintroduced by fluctuations
in climate, sea level, ocean circulation, and edaphic changes
associated with continued emergence of peninsular Flor-
ida. The trend, however, was toward pine, grass, rosette
palms, and scrub vegetation that could cope with physio-
logically dry deep sandy soil; the coast was fringed by trop-
ical mangrove communities.
Along the Atlantic coastal plain, sediments of Neogene
age are limited, but some data are available from drilling
operations that penetrated the Middle Miocene Kirkwood
Formation at Mays Landing, New Jersey (Owens et al.,
1988). Palynomorphs include abundant Pinus, Carya, and

Figure 7.4, facing page. Land form map of the conterminous United States with overlay of principal
Middle Miocene through Pliocene floras mentioned in the text: (1) Mays Landing, (2) Brandywine, (3) Martha's
Vineyard, (4) Ash Hollow, (5) Kilgore, (6) Teewinot, (7) Pyramid, (8) Mount Eden, (9) Chula Vista, (10) Sonoma, (11)
Trapper Creek, (12) Mascall, (13) Clarkia, (14) Succor Creek, (15) Trout Creek, (16) Stinking Water, (17) Ellensburg,
(18) Oak Grove, (19) Tulelake, (20) Deschutes, (21) Fingerrock, (22) 49-Camp, and (23) Latah (Thelin and Pike, 1991).
Table 7.1. Geographic distribution of Middle Miocene through Pliocene floras of North America.

Middle to Late Miocene (Including Mio-Pliocene)

Mid-Atlantic Kirkwood Owens et al. (1988)
Legler Greller and Rachele (1983), Rachele (1976)
Brandywine Groot (1991), McCartan et al. (1990)
Northeast Holmatindur Mudie and Helgason (1983)
Western Interior
Nebraska Ash Hollow Thomasson et al. (1990)
Kilgore MacGinitie (1962)
West
Wyoming Tweeinot Barnowsky (1984)
Nevada Pyramid Axelrod (1992a)
Verdi Axelrod (1958)
Fingerrock Axelrod (I992b)
Stewart Valley Axelrod (I992b)
Purple Mountain Axelrod (I992b)
49-Camp Axelrod (I992b)
Gillam Spring Axelrod (1992b)
Idaho Trapper Creek Axelrod (1964)
Clarkia Smiley and Rember (1981, 1985)
Smiley et al. (1975)
Manchester et al. (1991)
California Mount Eden Axelrod (1937, 1950c)
Oregon Succor Creek Cross and Taggart (1982), Fields (1990, 1992),Graham
(1963, 1965), Taggart et al. (1982)
Trout Creek Graham (1963, 1965)
MacGinitie (1933)
Stinking Water Chaney (1959), Chaney and Axelrod (1959)
Deschutes Chaney (1938)
Mascall Chaney (1959), Chaney and Axelrod (1959)
Washington Ellensburg Smiley (1963)
Latah Knowlton (1926)
Northwest
Alaska Cook Inlet Wolfe (1966, 1972), Wolfe and Hopkins (1967), Wolfe and
Leopold (1967), Wolfe et al. (1966)

Pliocene
Southeast
Florida Pinecrest Beds Willard (1993), Willard et al. (1993)
Northeast Martha's Vineyard Frederiksen (1984, 1989)
Kap K0benhavn Ovenden (1993)
West
Nevada Aldrich Station Axelrod (1956)
Chloropagus
Fallon
California Ricardo Webber (1933)
Furnace Creek Axelrod (1940)
Anaverde Axelrod (1950a)
Piru Gorge Axelrod (1950b)
Chula Vista Axelrod and Demere (1984)
Carmel Axelrod and Cota (1993)
Puente
Modelo
San Pablo Condit (1938)
Sonoma Axelrod (1944, 1950d)
Wildcat Dorf (1930)
Temblor Renney (1969)
Oregon Tulelake Adam et al. (1989,1990a,b)
Oak Grove Wolfe (1990)
Northwest
Alaska Lava Camp Hopkins et al. (1971)
Kivalina Hopkins et al. (1971)
Gubick Repenning and Brouwers (1992)
Ocean Point Nelson and Carter (1985, 1992)
 Middle Miocene through Pliocene History 249

Table 7.2. Palynomorphs from Pliocene Pinecrest Beds gesting cooler and more seasonal climates. The palynoflora
(Tamiami Formation), near Sarasota, Florida. of the Middle to Late Miocene Bryn Mawr Formation of
Arboreal Pollen Nonarboreal Pollen Spores
Maryland also suggests climates warmer than at present,
but cooling from earlier times (Pinus, Taxodium, Carya,
Juniperus Ephedra Isoetes Ilex, Quercus), and some remaining exotics (Sciadopitys,
Pinus Cyperaceae Lycopodium
Alangium, Engelhardia; Pazzaglia et al., 1997).
TCT Poaceae Equisetum
Acer Ambrosia Osmunda In New England, plant microfossils of Middle Miocene
Alnus Artemisia Polypodium and Pliocene age are known from Plymouth County and
Betula Aster Monolete spores Martha's Vineyard in eastern Massachusetts (Frederiksen,
Bumelia Asteraceae indet. Trilete spores 1984, 1989). The Middle Miocene vegetation was a rich,
Carya Chenopodiaceae Trilete, zonate spores warm-temperate mixed mesophytic forest with abundant
Castanea Ericaceae
Celtis Ilex conifers, while the Pliocene forest was less rich and more
Cercis Iva cool temperate (Table 7.4). Pollen of most exotics had dis-
Corylus Labiateae appeared from eastern North America by the Pliocene, giv-
Diospyros Leguminosae ing a more modern aspect to the vegetation. The absence of
Fagus Liguliflorae
Abies, Picea, and Betula in the Early Miocene Brandon Lig-
Fraxinus Polygonum
Juglans Ranunculaceae nite flora of Vermont attests to the Middle to Late Tertiary
Liquidambar Tubuliflorae cooling in northeastern North America since Brandon
Magnolia Umbelliferae time.
Myrica Evidence for climatic deterioration is further evident in
Nyssa
high latitude floras. The vegetation sequence in the Hol-
Ostrya-Carpinus
Platanus matindur Tuff floras of eastern Iceland (65° N; 10.3-9.5
Quercus Ma; Mudie and Helgason, 1983) is from a temperate hard-
Salix wood forest (Larix, Picea, Pinus, Sequoia, Taxodium,
Ulmus Carpinus-Ostrya, Carya, Fagus, Juglans, Nyssa, Quercus,
Adapted from Willard (1993). Ulmus, and others), to a cool-temperate deciduous hard-
wood-conifer forest (increase in Picea and Pinus, decrease
in Fagus), to a cold-temperate forest and subarctic Alnus
Quercus, with lesser amounts of TCT (pollen of the fami- woodland (abundant Picea, Betula sect. Nanae, Alnus
lies Taxodiaceae-Cupressaceae-Taxaceae, which cannot viridis; Mudie and Helgason, 1983). In the complex strata
be distinguished), Alnus, Betula, Fagus, Ilex, Liquidambar, of Banks Island (73° N) involving the Beaufort Formation
Momipites (Alfaroa-Engelhardia-Oreomunnea], Nyssa, and the Mary Sachs Gravel (Chapter 6; Fyles et al., 1994;
Sapotaceae, Tilia, Ulmus, and exotics such as Podocarpus Matthews, 1987; Matthews and Ovenden, 1990), the se-
and Pterocarya. Near the top of the section small amounts quence is from mixed deciduous-conifer forest (Picea,
of Cyperaceae, Gramineae, chenoam (pollen of the families Pinus dominant, Juglans eocinerea; Hills et al., 1974) to a
Chenopodiaceae-Amaranthaceae, which cannot be distin- decrease then disappearance of broad-leaved deciduous el-
guished), and Compositae-Asteraceae pollen appear, sug- ements and an increase in Larix and Picea. About 80
gesting local open habitats. species of Late Tertiary mosses are known from the Arctic
The overlying Cohansey Sand Formation is slightly islands (Banks, Ellesmere, Meighen, Prince Patrick, Queen
younger, but it is also Middle Miocene in age (estimated at Elizabeth Islands, Fig. 2.14; Beaufort Formation, Prince
-11 Ma; Greller and Rachele, 1983; Rachele, 1976). Pollen Patrick Island; Kap K0benhavn Formation, northern Green-
from the Legler Lignite lens is similar to that from the Kirk- land, Bennike and Bocher, 1990; Mary Sachs Gravel, Banks
wood Formation and includes Tsuga, as well as the exotics Island; Ovenden, 1993). They indicate a mosaic of diverse
Podocarpus, Alangium, Momipites, Pterocarya, Cyrilla, conifer forest, stunted conifer woodland, and depauperate
Gordonia, and Cyathea. forest-tundra vegetation. The most tundralike assem-
The Late Miocene Brandywine flora of southern Mary- blages are the youngest and are found in Plio-Pleistocene
land also reflects a warm-temperate climate (Groot, 1991; and Pleistocene deposits (-1.7 Ma; Repenning and Brouw-
McCartan et al., 1990; Table 7.3), as do the molluscan fau- ers, 1992). The Kap K0benhavn flora (82° 30' N; ~2 Ma) sug-
nas (Ward and Powars, 1989; Ward and Strickland, 1985). gests that the Arctic tree line was 2500 km north of its pre-
There is one extinct genus (Mneme, Lythraceae), two no sent position in northeastern North America (Funder et al.,
longer occur in Maryland (Sophora, west of the Appalachi- 1985), although it probably fluctuated with the tempera-
ans; Larix, south only to New Jersey), and four are exotic to ture changes described by Repenning and Brouwers (1992)
North America (the Asian Alangium, Pterocarya, Trapa, and summarized by Crowley and North (1991). The assem-
and Zelkova). Several thermophilic taxa present in the blage consists of remains similar to modern Larix occiden-
Legler flora are absent from the Brandywine assemblage talis, Picea mariana, Taxus, Thuja occidentalis, Betula,
(Cyathea, Podocarpus, Cyrilla, Gordonia, Momipites}, sug- and Cornus stolonifera. The mosses present, such as Schis-
250 Late Cretaceous and Cenozoic History of North American Vegetation

Table 7.3. Composition of Late Miocene Brandywine palynoflora, southern Maryland.

Isoetes Cornus Ptelea

Selaginella Euonymus Pterocarya
Ceratophyllum Fagus Quercus
Larix Gleditsia Sp. 1
Pinus Hamamelis? Sp. 2
Taxodium Ilex Sp. 3
Potamogeton Juglans Salix
Smilax Liquidambar Sambucus?
Typha Liriodendron Sophora
Acer Mneme Toxicodendron
Alangium Nuphar Trapa
Alnus Nyssa Ulmus
Ambrosia Parthenocissus? Vitis
Artemisia Platanus Zelkova type
Betula Populus
Carpin us-Ostrya Sp. 1
Carya Sp. 2
Celastrus Prunus

Adapted from McCarten et al. (1990).

tidium maritimum, are more typical of boreal forest than
open tundra. These were beginning to disappear from
northern Greenland by -2 Ma, and tundra was expanding
in Siberia (Wolfe, 1985). The flora bordered on an Arctic
Ocean without perennial sea-ice cover, as indicated by sed-
iments well sorted by wave action.
Fossil faunas of Tertiary age are extensively preserved in
midcontinent and western North America, and NALMAs
serve as the conventional temporal framework for describ-
ing the biotic history. In the Plains area the trend toward
lower temperatures and decreasing seasonal (winter dry)
precipitation continued, especially after -16 Ma. A mosaic
of open deciduous forest and grassland was present by the
Middle Miocene (Barstovian, -13 Ma) and was widespread
by the Late Miocene (Clarendonian-Hemphillian, -8 Ma).
The broad vegetation sequence was from deciduous forest
with prairie elements, to dry grassy woodland (savanna)
with riparian forests, to more extensive grassland at the
Miocene-Pliocene transition, but not yet pure prairie (Ax-
elrod, 1985; Leopold and Denton, 1987; Wolfe, 1985). Al-
though Elias (1942) proposed that the present Plains region
was grassland by the late Miocene, more recent studies
suggest it was savanna. At -10 Ma a major volcanic erup-
tion centered in southwestern Idaho deposited 3 m of ash
as far east as Nebraska and preserved the extensive Ash
Bed fauna. It includes camel, five species of horses, zebra,
and other warm dry-land browsing and grazing mammals.
At Big Springs the fauna by -2 Ma includes lemming,
shrew, and mastodon.
A number of fossil fruit and seed floras are known from
northern Texas to South Dakota, ranging in age from Hem-
ingfordian through Hemphillian (-18-4 Ma; Chaney and
Elias, 1938; Elias, 1932, 1935, 1942). Grasses and prairie
herbs became more abundant, widespread, and diverse
through the sequences. In northern Texas identifications of
Clarendonian plants include Juniperus, Arctostaphylos,
Celtis, Gymnocladus, Salix, Sapindus, and Ulmus; the as-
sociated fauna consisted of browsing and grazing horses,
oreodonts, and camels. In the panhandle of western Okla-
homa the fauna included tortoise, rhinoceros, peccary, and
an oreodont (Hesse, 1936). The estimated regional precipi-
tation was -510 mm (20 in), similar to the present, and rep-
resents the drier part of the Tertiary (Mio-Pliocene transi-
tion, 6-5 Ma; Axelrod, 1985). The area was sparsely
wooded with scattered open grasslands. Many of the older
grass identifications have been confirmed and augmented
by Thomasson (1978a,b, 1979, 1980a-c, 1982, 1983, 1984,
1985) using SEM and phytoliths. The Late Miocene Ash
Hollow Formation of Nebraska (6-7 Ma; Thomasson et al.,
1990) includes the sedges Carex, Cyperocarpus, and Ele-
ofimbris; the grasses Archaeoleersia, Berriochloa, Gramino-
phyllum, Nassella, and Panicum; the borages Biorbia and
Cryptantha; Celtis (Ulmaceae), Chenopodium (Chenopodi-
aceae), and Compositae-Asteraceae. Many of these floras
are small and preservation is often poor. The associated
faunas provide evidence of age, general vegetation type,
and environmental change (Clarendonian chronofaunal
decline; Webb, 1983) that is especially valuable for charac-
terizing the Neogene of the Plains region. It was noted in
Chapter 2 (CO2 concentration section) that at -6 Ma there
was a change in 13C in the plant debris from paleosols and
from the tooth enamel of ungulates that suggests a shift
from C3 plants (many trees and shrubs) to C4 plants (grasses;
Cerling et al., 1997). This, in turn, suggests a continued de-
cline in CO2 concentration, more pronounced seasonality
in rainfall, and lower temperatures.
 Middle Miocene through Pliocene History 251

Table 7.4. Miocene and Pliocene palynomorphs from east- A more extensive assemblage is the Late Miocene Kil-
ern Massachusetts. X = Taxa also present in the Pliocene. gore flora preserved in lacustrine sediments of the Valen-
tine Formation of northern Nebraska (Barstovian, 13-14
Miocene Pliocene
Ma; MacGinitie, 1962; Table 7.5). The most abundant com-
Sphagnum X ponents are megafossils of Platanus, Meliosma (Fig. 7.5A),
Lycopodium X Ulmus, Cedrela, Acer cf. negundo, and Robinia; and micro-
Selaginella fossils of Pinus, Carya, Ulmus (Fig. 7.5B), Sarcobatus(t;
Gleicheniidites or Neogenisporis greasewood), Liquidambar, Alnus (Fig. 7.5C), and Artemisia
(reworked?) X
(Fig. 7.5D). It is a streamside assemblage of broad-leaved de-
Osmunda
Ephedripites hungaricus ciduous forest bordered by a woodland of Pinus, Artemisia,
Abies Diospyros, Quercus, and Sarcobatus, and patches of grass-
Larix land (Axelrod, 1985). The climate was dry (estimated pre-
Picea X (?) cipitation 760-890 mm) and temperatures were declining,
Pinus X
but winter temperatures were still higher than at present as
Podocarpus
Sciadopitys indicated by crocodilians and large nonburrowing tortoises
TCT in associated fossil faunas (Hutchison, 1982) and by the
Tsuga Mexican exotics Cedrela, Cordia, and Meliosma.
Cyperaceae Along the eastern slopes of the southern Rocky Moun-
Gramineae X
tains, the woodland-grassland vegetation of the Plains
Sparganium-Typha
Alnus X graded into pinon-juniper woodland; at higher elevations
Betula X
Boehlensipollis hohlii (reworked?)
Carya Table 7.5. Composition of Late Miocene Kilgore flora of
Castanea northern Nebraska.
Chenopodiaceae -Amaranthaceae
Compositae
Megafossils
Short spine X
Long spine Chamaecyparis linguaefolia Platanus vitifolia
Comptonia Monocotyledonous leaf Populus crassa
Cornaceae Acer minor P. gallowayi
Corylus Carya libbeyi P. washoensis
Cyrilla-Cliftonia Cedrela trainii Prunus acuminata
Ericaceae X Celtis kansana Pterocarya oregoniana
Fagus Cladrastis prelutea Quercus argentum
Gordonia Cocculus rotunda Q. parvula
Ilex Cordia prealba Q. preturbinella
Juglans type Crataegus nupta Q. remington!
Liquidambar Diospyros miotexana Ribes infrequens
Ludwigia Fraxinus coulteri Robinia lesquereuxi
Malvaceae Mahonia marginata Ulmus speciosa
Momipites spackmanianus complex Meliosma predentata Vitis pannosa
(reworked?) Nyssa copeana
Morus
Myrica X (?)
NTicrofossils
Myriophyllum
Nyssa Algal spores Cedrela
Platanus Fungal spores Celtis
Pseudolaesopollis sp. (reworked?) Lycopodium Compositae
Pterocarya type Polyp od Ilex
Quercus Picea Juglans
Salix Pinus Liquidambar
Sapotaceae Cf. Sequoia Meliosma
Symplocos Cyperaceae Onagraceae
Tilia Gramineae Pterocarya
Ulm u s-Zelkova Acer Quercus
Umbelliferae Alnus Salix
Botrychium Ambrosia Cf. Sarcobatus
Palmae Artemisia Tilia
Carya (Paleogene type) Cf. Betula Cf. Ulmus
Caryophyllaceae(?) Carya
Adapted from Frederiksen (1984). Adapted from MacGinitie (1962).
252 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 7.5. (A) Meliosma predentata, Kil-
gore flora. (B) Pollen of Ulmus, Kilgore flora.
(C) Pollen of Alnus, Kilgore flora. (D) Pollen
of Artemisia, Kilgore flora (A-D). Reprinted
from MacGinitie with the permission of the
University of California Press. (E) Pseudot-
suga longifolia, Trapper Creek. Reprinted
from Axelrod (1964) with the permission of
the University of California Press.
it graded through Quercus, Artemisia, and various herbs
into a western montane coniferous forest. The Teewinot
Lake flora at Jackson Hole, Wyoming (late Clarendonian,
~8 Ma; Barnowsky, 1984), shows a shrubland to near desert
with Ephedra and Sarcobatus, through which riparian veg-
etation extended (Carya, Pterocarya, Salix, Ulmus-
Zelkova], and a western montane coniferous forest of
Abies, Picea, Pinus, and Tsuga at higher elevations. Asian
exotics had nearly disappeared from the eastern slopes of
the Rocky Mountains by Clarendonian time, but they per-
sisted later west of the Rocky Mountains. In the western
Plains and adjacent foothills, as in the central Plains,
grasses were present but not widespread or abundant in
the Middle and Late Miocene.
The central and northern Rocky Mountains became a
greater barrier between vegetation to the east and west
during Middle Miocene through Pliocene time. Only six
species from the Kilgore flora occur in floras of similar age
of the Columbia Plateau (Leopold and Denton, 1987). Fos-
sil grasses from the Plains have affinities mostly to the east
and south into Mexico, while those west of the Rocky
Mountains have affinities to the north.
To the south and west of the Rocky Mountains in the
southwestern United States the Miocene vegetation was
mostly a pifion pine-juniper woodland and shrubland
with evergreen oak, grading upward into depauperate west-
ern montane coniferous forest. There were exceptions to
the dry vegetation, which were created by local topo-
graphic lows across the Sierra Nevada. The Pyramid flora
in western Nevada (15.6 Ma; Axelrod, 1992a) is to the west
of the Buffalo Canyon, Middlegate, and Eastgate floras dis-
cussed earlier (Chapter 6). It is dominated by deciduous
hardwoods (Acer, Alnus, Betula, Cladrastis, Fraxinus, Os-
trya, Platanus, Populus), including members of a flood-
plain association (Taxodium, Quercus simulata). The esti-
mated MAT is ~13.5°C (CLAMP 7.5° C), annual mean range
15°C, annual precipitation 900 mm (35-40 in.), and the
paleoelevation -550-600 m (1800-2000 ft; presently 1.4
km). However, new calculations using leaf physiognomy
and mean annual enthalpy suggest higher paleoelevations
from many Middle Miocene floras in the Basin and Range
Province (e.g., 2.8 km for the Pyramid flora; Wolfe et al.,
1997). Recall from Chapter 1 (Pacific coastal and Rocky
Mountains coniferous forests) that there is a modern ana-
 Middle Miocene through Pliocene History 253

log for the effect of topographic lows across the western Table 7.6. Composition and paleocommunities of Mount
mountains on vegetation. A similar opening along which Eden flora.
the Columbia River flows extends today through the Cas-
cade Mountains, allowing heavy rains and cold tempera- Desert
tures to penetrate onto the western slopes of the Rocky Baccharis beaumontii Eysenhardtia pliocenica3
Mountains near the Columbia Plateau. These forests share Cercidium edenensisa Ficus edenensisa
species that elsewhere are used to define different montane Chilopsis sp. Forestiem buchananensis
coniferous forest associations. The Pyramid flora illustrates Condalia coriacea Prunus (Emplectocladus)
Dodonaea californicaa preandersonii
one aspect of the importance of local topography on the Ephedra sp. P. (Emplectocladus} prefremontii
composition of western North American paleovegetation.
Younger Mio-Pliocene floras in the region record the
Shrubland-Chaparral
partly orographic decrease in rainfall and the development
of increasingly arid vegetation from the rise of the Sierra Arctostaphylos preglauca Eysenhardtia pliocenicaa
A. prepungens Ficus edenensisa
Nevada and coastal mountains (Aldrich Station, 12.4-13.3 Ceanothus edenensis Fraxinus edenensis
Ma; Chloropagus, 14.3 Ma; Fallen, 15.4 Ma; Axelrod, 1956; C. precuneatus Prunus (Laurocerasus)
Wolfe et al. 1997; see also Axelrod, 1958, 1980, for index C. prespinosus mohavensis
maps of the many Neogene floras in Nevada and Califor- Cercidium edenensisa Quercus pliopalmeri
nia). The four florules of the Purple Mountain flora of west- Cercocarpus cuneatus Rhamnus moragensis
Dodonaea californicaa Rhus (Malosma) prelaurina
ern Nevada (~14.8 Ma) grew under an estimated rainfall of R. (Schmaltzia) moragensis
760-890 mm (30-35 in.; oldest), decreasing to 625-760
mm (25-30 in.; Axelrod, 1995). The Verdi flora to the west
Woodland
of Middlegate in western Nevada consists of flood-plain
vegetation, with oak woodland, chaparral, and savanna on Pinus pieperi Quercus convexaa
Arbutus prexalapensis Q. dayanaa
the slopes, and pine and fir in the bordering hills. Rainfall Eysenhardtia pliocenica3 Q. douglasoides
in the lowlands is estimated at 450-500 mm (18-20 in.; Juglans beaumontii Q. lakevillensisa
Axelrod, 1958). J. nevadensis Salix edenensisa
Several floras of Pliocene age are known from the arid Persea coalingensis S. truckeanaa
region of interior southern California (Furnace Creek: Axel- Platanus paucidentataa S. wildcatensisa
Populus sonorensis Sapindus oklahomensis
rod, 1940; Anaverde: Axelrod, 1950a; Piru Gorge: Axelrod,
1950b; Ricardo: Webber, 1933). During parts of the Pliocene
the Central Valley of California was a marine embayment Western Montane Coniferous Forest
(Stanton and Dodd, 1970), and the Gulf of California ex- Cupressus preforbesii Amorpha oblongifolia
tended further north than at present. One of the most exten- Pinus hazeni Philadelphus nevadensis
P. pretuberculata Populus pliotremuloides
sive assemblages is at Mount Eden (-5.3 Ma) in the San Jac- Pseudotsuga premacrocarpa Quercus dayanaa
into Mountains southeast of Los Angeles (Axelrod, 1937, Amorpha oblongifolia
1950c; Table 7.6). It was then -30 km east of the marine em-
Adapted from Axelrod (1950c).
bayment in the Salton Sink Basin (Axelrod and Cota, 1993). a
These species occur in more than one unit.
The species comprise six communities that reflect semiarid
conditions and low to moderate elevations in the interior.
The near-desert community included members typical of The Mediterranean (winter-wet) climate that presently
the more mesic parts of the present Sonoran Desert. The characterizes southern coastal California is evident from
shrubland elements are mostly holdovers from the subtrop- the N-S precipitation cline. Annual rainfall at Mt. Baker
ical vegetation that already characterized interior southern in northern Washington is 2818 mm (110.96 in.), at Eureka
California during the Miocene (e.g., Tehachapi flora). Ele- (northern California) it is 954 mm (37.58), at San Francisco
ments available to occupy these habitats were, in turn, pre- (-central California) it is 514 mm (20.23), and at San Diego
sent as local, edaphic and slope-controlled, preadapted (southern California) it is 257 mm (10.11). In San Diego the
types since Middle Eocene Green River time. In the winter months receive 46.7 mm (1.84 in.) of precipitation
Pliocene these were coalescing and expanding in response in December, 50 mm (1.97) in January, 56 mm (2.22) in Feb-
to drying climates. Rainfall is estimated at 300-450 mm ruary, and 40 mm (1.59) in March, or 75% of the annual
(12-18 in.), but with some summer rain, in contrast to the precipitation. In Tucson, Arizona, to the west and at about
summer-dry conditions that developed later in the Pliocene the same latitude (32° N), the annual precipitation is simi-
and Pleistocene. It is likely that at least some elements of lar (283 mm; 11.16 in.); but the winter months receive (26
the present dry vegetation of southern coastal California mm; 1.02 in), (21 mm; 0.81 in), (23 mm; 0.89 in), and (19
(Diegan sage; Axelrod, 1978) spread from the arid interior. mm; 0.76 in.) or 34% of the annual precipitation.
The California grassland becomes recognizable as a distinct The explanation for Mediterranean climates and sclero-
association in the Pleistocene. phyllous vegetation in southern California involves a com-
254 Late Cretaceous and Cenozoic History of North American Vegetation
plex of factors. One component are summer storms off the
northeast Pacific Ocean that move progressively southward
and do not reach southern California until the winter sea-
son. This pathway is a result of the coastal mountains that
deflect east-trending storm tracks to the south. This places
the origin of Mediterranean climates late in the Tertiary
(Late Pliocene and Pleistocene) by which time the coastal
mountains had reached substantial heights. Another com-
ponent is the presence of the southern Rocky Mountains,
the Sierra Nevada, and the high pressure arm of the Hadley
regime that prevent rain-generating summer lows off the
Caribbean Sea and Gulf of Mexico from penetrating to the
west coast. Elements of dry sclerophyllous vegetation, al-
ready present in western North America in the Eocene
(e.g., Green River flora), coalesced after the Middle to Late
Miocene drying trend and were available to occupy the
Mediterranean-type habitats of southern California by the
Pliocene.
The broad temperature trend during the Miocene along
the southern California coast is reflected in the 13C content
of hopane and sterane compounds preserved in the Mon-
terey Formation at Naples Beach (Schoell et al., 1994;
Chapter 3). The decreasing 13C values parallel increasing
amounts of 18O and show a prominent cooling at -15 Ma.
Coastal waters changed from well mixed to a highly strati-
fied photic zone (upper 100 m) as at present. The decrease
in temperature in the zone between the Early and Late
Miocene is estimated at 3.1-4.3°C. Near the end of the
Miocene the lower photic-zone temperature was near the
present 10°C. One effect of these cooling coastal waters,
comparable to that of the cold Humboldt Current observed
today along western South America (Chapter 2), was re-
duced precipitation along the borderlands. The 13C-based
cooling trend reversed briefly at 7-8 Ma, consistent with
the warming shown on the 18O-derived benthic tempera-
ture curve (Chapter 3).
On the California coast near the Mexican border a small
flora is preserved in the San Diego Formation at Chula Vista
(~3 Ma; Axelrod and Demere, 1984). It includes Pinus
diegensis (cf. the modern P. radiata), P. jeffreyoides (P. jef-
freyi), P. pieperi (P. sabiniana, digger pine), Palmae, Persea
coalingensis (avocado), Populus alexanderi (P. tricho-
carpa), Salix wildcatensis (S. lasiolepis), Platanus pauci-
dentata (P. racemosa), and Quercus lakevillensis (Q. agrifo-
lia, California live oak). It is a pine-oak woodland with
palms, avocado (Persea), cottonwood, willow, and syca-
more along the streams. Environments were more mesic
than in the interior, as shown by other floras in the region
(Carmel: an exotic terrane flora, Puente, Modelo; Axelrod
and Cota, 1993). The MAT is estimated at 16°C (presently
12-13°C), and the annual precipitation in the lowlands
was 500-580 mm (20-23 in.) compared to the present 220
mm (9 in.). The flora grew during the mid-Pliocene warm
period, and a MAT 3-4°C warmer than at present brought
more summer rain. The summer-dry winter-wet Mediter-
ranean climate (Arroyo et al., 1995) that presently charac-
terizes the area developed in the Late Pliocene and Pleis-
tocene.
In the San Francisco Bay area several floras span the in-
terval between the Mio-Pliocene and the Late Pliocene
(e.g., Sonoma: Axelrod, 1944, 1950d; San Pablo: Condit,
1938). These are west of the Sierra Nevada and received
moisture from the Pacific Ocean. The Sonoma flora in-
cludes several florules between Napa and Santa Rosa about
50 km to the northwest. The Napa florule is slightly more
interior and represents a pine (cf. P. ponderosa)-Douglas
fir (Pseudotsugo) forest near sea level and an oak woodland
(Quercus dayana, Q. wislizenoides)-chaparral (Arc-
tostaphylos fergusoni, Cercocarpus cuneatus, C. lineari-
folius, Photinia sonomensis) on the slopes, together with
several elements that now border the redwood forest on
drier sites (Amorpha condoni, Holodiscus harneyana). At
Santa Rosa the more coastal vegetation included better rep-
resentation of the modern redwood forest (Sequoia affinis,
Alnus rubroides, Castanopsis sonomensis, Ceanothus
edensis, Lithocarpus klamathensis, Mahonia simplex,
Salix boisiensis, Umbellularia salicifolia) and a higher ele-
vation coastal coniferous forest of Abies sonomensis, Picea
sonomensis, Tsuga sonomensis, and associated Rhododen-
dron sp. and Vaccinium sonomensis. Rainfall is estimated
at 1000 mm (40 in., 280 mm more than at present; Axelrod,
1944) with greater summer rain and a ~3°C warmer MAT.
Wolfe (1990) estimates rainfall at 100 mm higher and a
2-4°C warmer MAT. Other fossil floras in the region show-
ing similar vegetation include the Wildcat (Dorf, 1930), San
Pablo (Condit, 1938), and Temblor (Renney, 1969).
To the north around the Columbia Plateau in the Middle
to Late Miocene a rich broad-leaved deciduous forest grew
at low to midelevations (below ~600 m; Metasequoia, Tax-
odium, Ailanthus, Betula, Cedrela, Diospyros, Fagus,
Hamamelis, Ilex, Magnolia, Persea, Phoebe, Pterocarya,
Quercus, Sassafras, Ulmus, Zelkova). This graded through
a mixed conifer-hardwood forest (Abies, Chamaecyparis,
Picea, Pinus, Acer, Betula, Quercus, Ulmus, Zelkova) into
a western montane coniferous forest at the highest eleva-
tions. The Tertiary floras of the Columbia Plateau are
among the richest and most extensive in the world. The
forests grew under warm-temperate summer-wet condi-
tions and are prominent until -4.5 Ma (Hemphillian), after
which progressively arid summer-dry conditions devel-
oped to the lee of the rising Sierra Nevada and Cascade
Mountains. The vegetation was part of a broad zone of de-
ciduous forest and mixed conifer-hardwood forest that ex-
tended into Alaska (Seldovia Point flora) and across the
midlatitudes of the northern hemisphere into Europe and
Asia. Throughout the sequences, pollen of herbs and xeric
elements are present but not yet abundant.
The Trapper Creek flora of south central Idaho, earlier
estimated to be 15-16 m.y. in age (Axelrod, 1964), is now
placed at 10.5-12 m.y. (Clarendonian; Armstrong et al.,
1975; Fields, 1983). It is a summer-wet deciduous forest
with some broad-leaved evergreens and a montane conifer-

Table 7.7. Suggested taxonomic and geographic affinities of common members of late Middle Miocene Stinking Water
flora, Harney County, Oregon.
Fossil East America
Glyptostrobus oregonensis
Alnus harneyana (Fig. 7.10C)
Alnus relatus A. maritima
Platanus dissecta P. occidentalis
Populus lindgreni (Fig. 7.10B) P. heterophylla
Quercus dayana (Fig. 7.10A) Q.virginiana
Q. hannibali
Q. prelobata (Fig. 7.10D)
Q. pseudolyrata Q. borealis
Q. simulata
Ulmus speciosa U. fulva
Adapted from Chaney (1959).
hardwood forest. Components of the deciduous forest in-
cluded swamp forest (Taxodium, Nyssa), lake border vege-
tation (Acer, Populus, Salix), valley forest (Alnus, Ame-
lanchier, Carya, Cornus, Fraxinus, Parthenocissus, Prunus,
Pterocarya, Sophora, Ulmus; broad-leaved evergreens Ber-
beris, Ilex, Quercus), and mountain slope vegetation (Pinus,
cf. P. ponderosa, P. monticola; Calocedrus, Cedrus, Garrya,
Rhus). Some frost-sensitive plants (e.g., Liquidambar)
began disappearing in the Middle Miocene, although Tax-
odium swamps persisted in Idaho until -12 Ma and in
eastern Washington until ~8 Ma (Leopold and Denton,
1987). Conifers included Abies, Keteleeria, Picea, Pseudot-
suga (Fig. 7.5E), Sequoiadendron, and Tsuga. The flora sug-
gests higher and cooler conditions than do other floras in
the region (Succor Creek, Stinking Water, see below), and
the dominant vegetation was coniferous forest; the broad-
leaved community is interpreted as mostly recovery vege-
tation following disruption by at least five episodes of vol-
canic activity (Taggart and Cross, 1990).
The Mascall flora is from along the East Fork of the John
Day River in northeastern Oregon, and it is Early Barstov-
ian in age (-16 Ma). The most abundant megafossils (>1%)
in order are Taxodium dubium, Quercus pseudolyrata,
Carya bendirei, Quercus dayana, Platanus dissecta, Quer-
cus merriami, Acer bolanderi, Metasequoia occidentalis,
Ginkgo adiantoides, Ulmus speciosa, Acer minor (A. medi-
anum), Cedrela trainii (C. pteraformis), Ulmus pauciden-
tata, Betula thor, and Acer scottiae. The flora reflects a
swamp cypress and deciduous forest in the lowlands and
on the slopes and a hardwood-conifer forest in the up-
lands (Chaney, 1959; Chaney and Axelrod, 1959). Affinities
are with the present vegetation of eastern North America,
western North America, and eastern Asia. (For a complete
listing of a Columbia Plateau flora, see the more recently
studied Succor Creek flora described below; examples of
taxonomic and geographic affinities of some Columbia
Plateau fossils with modern species are given in Table 7.7.)
To the north in Idaho, the Clarkia flora (-16 Ma), about
Middle Miocene through Pliocene History 255
Similar Modern Species
East Asia West America
G. pensilis
A. hirsuta
A. japonica
P. mcemosa
P. rotundifolia
Q. chrysolepis
Q. lobata
Q. kelloggii
Q. stenophylla, Q.myrsinaefolia
55 mi northeast of Moscow, is preserved in lacustrine de-
posits resulting from damming of the drainage system by
lava flows to form the proto-St. Maries River and Clarkia
Lake. Nearby basalts and numerous ash layers (2 mm to 50
cm thick in the section) indicate frequent volcanic activity
similar to that at Trout Creek (Figs. 3.10, 3.11); the effects
were probably similar to those described below for the
Succor Creek flora. The systematics of the Clarkia flora
have not been presented in detail and identifications are at
the generic level. Three communities are represented (Smi-
ley and Rember, 1981,1985; Smiley et al., 1975): a bottom-
land swamp and riparian forest (Taxodium, Chamaecy-
paris, Populus, Salix, Nyssa), a mesic slope forest
(Metasequoia, Sequoia, Thuja, Acer, Aesculus, Alnus, Be-
tula, Carya, Castanea, Cercidiphyllum, Cornus, Corylus,
Diospyros, Fagus, Fraxinus, Hamamelis, Juglans, Liq-
uidambar, Liriodendron, Ostrya, Paulownia, Pterocarya,
Quercus, Sassafras, Ulmus, Zelkova), and a drier slope for-
est (Pinus, Amorpha, Amelanchier, Celtis, Crataegus,
Quercus, Rhus, Rosa). A western montane coniferous for-
est was also present in the vicinity as shown by the pres-
ence of pollen and megafossils of Abies and pollen of Picea
and Tsuga. The extinct genera Nordenskioldia (infmctes-
cences, fruits; widespread in the Paleogene of the Northern
Hemisphere) and Zizyphoides (leaves) have been reported
from the Clarkia flora (Manchester et al., 1991). The flora
also contains fossil infructescences and fruits of Tro-
chodendron (Manchester et al., 1991). The principal vege-
tation association was a mixed mesophytic forest of the de-
ciduous forest formation.
Ancient DMA
Leaf compressions of unoxidized cuticular and mesophyll
tissue in the Clarkia flora contain original pigmentation.
This imparts to the assemblage a feature rare in fossil flo-
ras: the potential preservation of ancient DNA and other
organic compounds (Golenberg, 1991; Golenberg et al.,
256 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 7.6. (A, right) Succor

Creek locality, southeastern
Oregon. (B) Glyptostrobus
oregonensis,vegetative branch.
(C) Glyptostrobus oregonensis,
cone. (D) Fraxinus coulteri.
(E) Sassafras columbiana.
(F) Hydrangea-like calyx.
(G) Ulmus speciosa. (H) Acer
chaneyi (A. benderi in Gra-
ham, 1965; see Wolfe and
Tanai, 1987). (I) Pollen of
Picea. (J) Pollen of Abies.
(K) Pollen of Ambrosia.
(L) Pollen of Chenopodiaceae-
Amaranthaceae. (M, N) Pollen
ofPinus. (O) Pollen of Carya.
(P) Pollen ofjuglans. Adapted
from Graham (1965).

1990; Soltis and Soltis, 1993; Soltis et al., 1992). A partial
sequence [759 base pairs (bp)] was obtained from the
chloroplast DNA rbcL gene from tissues that were part of
an extinct species of Magnolia (M. latahensis). The gene
encodes for ribulose 1,5-bisphosphate carboxylase/oxyge-
nase, which is an enzyme responsible for fixing atmo-
spheric carbon. Comparisons with rbcL sequences of mod-
ern Magnolia species revealed affinities with that genus. In
addition, there were differences between the modern and
extinct species of 17 base substitutions of which 12 were
transitions and 13 were silent third-position substitutions
(Soltis and Soltis, 1993). Similar analyses were made of
fossil Taxodium from the Clarkia flora and compared with
modern Taxodium, Metasequoia, and Pseudotsuga (Soltis
et al., 1992). These suggested the ancient DNA was from
Taxodium. Amplified products (780 and 1380 bp) were
larger than achieved with ancient animal DNA, which
were interpreted as a consequence of less degradation
under the favorable preservation conditions at Clarkia.
DNA sequences have also been obtained from fossil Ju-
niperus and Symphoricarpus (snowberry, Caprifoliaceae)
from packrat middens older than 45,000 yrs (Rogers and
Bendich, 1985) and from Oligo-Miocene bee and termite
remains preserved in Dominican amber that are -22 Ma in
age (Cano et al., 1992; DeSalle et al., 1992). However, a new
study based on the extent of racemization of aspartic acid,
alanine, and leucine suggests that chloroplast DNA from
the Clarkia material has undergone extensive microbial de-
gredation (Poinar et al., 1996), and another study (Austin et
al., 1997) casts doubt on other reports of ancient DNA from
insects preserved in amber.
The preservation quality of the Clarkia specimens has
allowed another innovation in the study of fossil leaves.
Measurements of stomatal guard cells have been used to
estimate chromosome number and ploidy level of the fossils
(Platanaceae, Magnoliaceae, Lauraceae; Masterson, 1994).
The Succor Creek flora (Fig. 7.6) of Barstovian age (14-15
[14.8-15.5] Ma; Downing and Swisher, 1993) in southeast-
ern Oregon (Malheur County) and adjacent Idaho (Owyhee
County) was deposited at ~1000 m (Cross and Taggart,
1982) with greater highlands to the south. The flora has
been studied extensively (Brooks, 1935; Cross and Taggart,
1982; Fields, 1992; Graham, 1963, 1965; Smith, 1938,1939;
Taggart et al., 1982) and presently consists of -175
megafossil and -100 microfossil taxa for a total of -275
named taxa (Table 7.8). However, when parallel names for
different organs of the same taxon are taken into account, a
more realistic number is -160 species (P F Fields, personal
communication, 1996). It is one of the largest fossil floras
known for North America. About 25-30% of the species
have entire-margined leaves (temperate climate), although
the number of specimens required for leaf physiogonomic
analysis (30 dicot leaves) and their percentage varies
among the different localities.
The principal communities were warm-temperate forest
with evergreen elements in the lowlands (Cedrela, Maho-
nia, Oreopanax, Quercus] and progressively cool-temper-
ate deciduous forest, mixed conifer-hardwood forest, and
western montane coniferous forest at higher elevations.
Lake margin, marsh, and riparian vegetation bordered the
depositional basin (Fig. 7.7). The effects of frequent vol-
canic activity are evident in plant microfossil assemblages
from closely spaced samples. At the Whiskey Creek section

Figure 7.6. continued
of the Succor Creek Formation, the interval between sam-
ples 13 and 14 reveals a significant disruptive event. The
forest abruptly declines and is replaced by a herbaceous
vegetation of Gramineae, Amaranthaceae-Chenopodi-
aceae, Compositae-Asteraceae, and Malvaceae. This is in-
terpreted as a postdisruption sere (successional stage) fol-
lowing destruction of the forest by volcanic ash fall. The
succession progressed intermittently through a pine park-
land back to deciduous forest. Mats of conifer needles as-
sociated with charcoal may reflect direct blast effects, out-
gassing, or heat. The volcanism that created conditions
favorable to fossilization (Chapter 2, Figs. 2.21, 2.22) also
periodically set succession back in local sites to bare
ground, comparable to that witnessed in the May 18, 1980
eruption of Mount St. Helens. Recognition of these events
Middle Miocene through Pliocene History 257
is important in distinguishing between cyclic vegetational
changes due to climate and those resulting from succession
and repeated regional catastrophies.
In an early demonstration that organic compounds such
as flavonoids do preserve for millions of years in plant fos-
sils, Niklas and Giannasi (1977; Niklas, 1981) recovered
kaempferol, dihydrokaempferol, an n-alkane chain length
range of 10-32 carbons, hydroxy acids, steranes, triter-
penoids, and methyl pheophorbide a from Zelkova orego-
niana leaves from Succor Creek (Fig. 7.8).
The Trout Creek flora in Harney County, southeastern
Oregon, is preserved in diatomite with numerous lenses of
volcanic ash (Figs. 3.10, 3.11; Graham, 1963, 1965; Mac-
Ginitie, 1933). It is slightly younger (13.8 Ma) than the Suc-
cor Creek flora, as reflected by the absence of the Mexican
 258 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 7.6. continued

exotics Cedrela and Oreopanax and the Asian Ginkgo, 1996) in some facies. The suggested affinities of these fos-
Glyptostrobus, and Pterocarya. It also grew under cooler sils with living species (fide Chaney, 1959) are listed in
conditions, because in addition to the absence of the warm Table 7.7. The distribution of the modem species illus-
temperate to subtropical Cedrela and Oreopanax, conifers trates the broad geographic occurrence of temperate plants
are more diverse and abundant at Trout Creek than those of across the midnorthern latitudes during the Middle Ter-
the lower paleoelevations in northern Succor Creek locali- tiary and the relictual nature of many present-day taxa.
ties. However, they do compare well with conifers from the The Ellensburg flora of south-central Washington is -14
upper paleoelevations from southern Succor Creek sites. m.y. in age (Axelrod, 1992b; Evernden and James, 1964;
Conifers are rare in the Stinking Water flora (11 Smiley, 1963; [10.5-13.5, depending on the horizon]), and
[12.5-13.4] Ma) just to the north in Harney County. This is preserves a piedmont vegetation from the eastern flank of
a lowland flora with abundant oaks, together with Glyp- the Cascade Range. The flora consists of a series of florules
tostrobus oregonensis, Alnus harneyana, A. relatus, Pla- that record an early lowland swamp cypress forest (Tax-
tanus dissecta, Populus lindgreni, and Ulmus speciosa. odium dubium), a midaltitude mixed deciduous hardwood
Swamp habitats are evident by the abundant fruits of forest (Acer columbianum, Betula thor, Fagus washoensis,
Trapa americana and rootstocks of Nymphaeites nevaden- Fraxinus coulteri, Platanus dissecta, Quercus bretzi, Sas-
sis (Nymphaea, fide E E Fields, personal communication, safras hesperia, Ulmus newberryi], and a younger post-
Table 7.8. Composition of Late Miocene Succor Creek flora of southeastern Oregon.
Megafossils Microfossils
Equisetum octangulatum Gymnocladus dayana Alternaria Carpinus
E.sp. Hiraea knowltoni Eumycophyta Carpinus-Ostrya
Davallia solidites Hydrangea bendirei Chytridaceae Carya (Fig. 7.6O)
Osmunda claytonities H.-like calyx (Fig. 7.6F) Fungi Imperfecti Castanea
Polypodium sp. Ilex fulva Fungal remains (aff. uncertain) Celtis
Woodwardia deflexipinna Juglans sp. Micrhystridium Chenopodiaceae -
Abies sp. Legume (fruit, leaf) Psophosphaera Amaranthaceae (Fig. 7.6L)
Cephalotaxus californica Legume (thorny branches) Sigmapollis Compositae
Cupressus-Juniperus Liquidambarsp. Acritarchs undifferentiated Low spine
Ginkgo adiantoides Lithocarpus klamathensis Botryococcus High spine
Glyptostrobus oregonensis Magnolia ovulata Ovoidites Corn us
(Fig. 7.6B, C) Mahonia macginitiei Pediastrum Corylus
Keteleeria sp. M. malheurensis Algal remains (aff. uncertain) Eleagnaceae
Librocedrus masoni M. reticulata Monolete spores EpilobiumC?)
Metasequoia occidentalis M. simplex Trilete spores Ericaceae
Picea lahontense M. trainii Lycopodium Fagus
P. magna M. sp. Equisetum Fraxinus
Pinus harneyana Nymphaea sp. Davallia Ilex
P. sp. (leaf) Nyssa copeana Osmunda Juglans (Fig. 7.6P)
P, sp. (cones) N. hesperia Polypodium Leguminosae (polyad)
Pseudotsuga longifolia Oreopanax precoccinea Woodwardia Liquidambar
Sequoia sp. Ostrya oregoniana Unknown fern spore Lithocarpus
Taxodium sp. Persea pseudocarolinensis Ephedra Mahonia
Thuja dimorpha Photinia sp. Abies (Fig. 7.6J) Malvaceae
Tsuga sonomensis Platanus bendirei Cedrus Nymphaea
Carex sp. P. youngii Cupressaceae Nymphaeaceae
Cyperites sp. (Gramineae) Populus eotremuloides Keteleeria Nyssa
Smilax sp. P. lindgreni Picea (Fig. 7.61) Onagraceae
Typha lesquereuxi P. payettensis Pinus (Fig. 7.6M.N) Ostrya
Acer busamarum busamarum P. pliotremuloides Podocarpus Pachysandra
A. b. fingerrockense P. voyana Pseudots uga-Larix Platanus
A. chaneyi (Fig. 7.6H) P. washoensis Taxaceae Populus
A. latahense Ptelea miocenica Taxodiaceae Pterocarya
A. medianum Pterocarya mixta TCT Quercus
A. negundoides Prunus sp. Tsuga Rosaceae
A. schorni Pyrus mckenziei Gramineae Salix
A. scottiae Quercus dayana Potamogeton Sarcobatus
A. septilobatum Q. eoprinus Typha Shepherdia
A. tyrellense Q. hannibali Acer Tilia
Ailanthus indiana Q. prelobata Alnus Ulmus
Alnus hollandiana Q. pseudolyrata Ambrosia (Fig. 7.6K) Umbelliferae
A. relatus Q. simulata Artemisia Zelkovatf)
A. sp. Q. subsp. erythrobalanus Betula
Amelanchier couleeana Q.sp. Caprifoliaceae
Anoda suckerensis Rhus sp.
Arbutus idahoensis Ribes sp.
A. trainii Salix hesperia
Betula thor S. succorensis
B. sp. (leaf, wood) S. sp.
B. sp. (seed) Sassafras columbiana
Carpinus sp. (Fig. 7.6E)
Carya sp. Symph oricarpos
Castanea spokanensis salmonensis
Castanopsis sp. Tilia aspera
Cedrela pteraformis Ulmus knowltoni
Celtis sp. U. newberryi
Com us ovalis U. owyhensis
Crataegus gracilens U. paucidentata
C. sp. U. speciosa (Fig. 7.6G)
Diospyros oregoniana Umbellularia sp.
Evodia sp. Vaccinium sonomensis
Fagus washoensis Zelkova browni
Fraxinus coulteri (Fig. 7.6D)

Adapted from Fields (1990). For Amelanchier see Schorn and Gooch (1994); for Ptelea see Call and Dilcher (1995). A revised list based on Fields (1996) is
in preparation (R E Fields, personal communication, 1996).
 260 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 7.7. Generalized re-

construction of climax pale-
ovegetation, Succor Creek,
southeastern Oregon.
Reprinted from Cross and
Taggart (1982) with the per-
mission of Aureal Cross,
Ralph Taggart, and the Mis-
souri Botanical Garden.

Ellensburg semiarid oak forest (Quercus prelobata; small-
leaved Ceanothus precuneatus, Ilex opacoides, Ulmus
moorei, U. paucidentata). The climate changed from a
humid summer-wet to a semiarid summer-dry regime. The
Ellensburg and associated floras document that through
-10-8 Ma the vegetation was not yet affected by decreas-
ing moisture from the rise of the Cascade Mountains. The
Latah flora at Spokane, Washington-Coeur d'Alene, Idaho
(15.8 Ma), has not been revised since Knowlton's (1926)
study, but it also represents a mixed mesophytic forest.
Pollen from a series of strata in the Idaho Group of the
Snake River Plain record the decline of the deciduous forest
from the Late Miocene through the Pliocene (Leopold and
Wright, 1985). The older assemblages include Carya,
Juglans, Quercus (megafossils), Pterocarya, and Ulmus,
along with diverse woods of deciduous hardwoods. Nyssa
disappeared from western North America after the Late
Miocene (Eyde and Barghoorn, 1963; Wen and Stuessy,
1993). Near the end of the sequence (-3-2 Ma) the flora is
impoverished pine and other conifers, and hardwoods are
rare. In still younger Plio-Pleistocene floras, components of
a Great Basin desert-shrubland with grasses (to 60%; Sar-

Figure 7.8. Pigmented leaf

of -14-15 m.y. old Zelkova
oregoneniana, Succor Creek.
Photograph courtesy of
Karl J. Niklas.

cobatus, Artemisia) increase, and there is further decline of
the deciduous forest (trace amounts of Carya and Ulmus).
Fish faunas and oxygen isotopic composition of fish otoliths
from the Snake River Plain record a change from warm
moist with cool summers and mild winters (Late Miocene)
to cooler summers (Smith and Patterson, 1994). In the Great
Salt Lake Basin of northern Utah, pollen from an Amoco
Production core revealed the vegetation of the eastern Great
Basin for the last ~4 Ma (Davis, in press). Cold-desert shrubs
of the Chenopodiaceae (Amaranthus or Sarcobatus) were
present on the slopes, and Cyperaceae (sedges) and Typha
(cattail) were along the streams and lake margin. Only a
small amount of pollen from a few holdovers of the decidu-
ous forest was recovered (Juglans, Ostrya-Carpinus, Shep-
herdia, Ulmus). The Early Pliocene climate is estimated at
3-4°C warmer and 50-100 mm drier than at present. Esti-
mates for the Oak Grove Fork flora of northwestern Oregon
(~3 Ma) are for temperatures 2—4°C higher and precipitation
1000 mm lower (Wolfe, 1990). Pollen records reveal abun-
dant conifers and more mesic plants (Abies, Cedrus, Alnus)
than do samples of similar age inland, suggesting greater
continentality toward the interior.
In northern California near the Oregon border a core
from Tulelake, just to the east of the southern Cascade
Range, includes a Pliocene section (3-2 Ma; Adam et al.,
1989, 1990a,b; Fig. 7.9). The lower portion of the core re-
flects a mixed conifer forest similar to that on the lower
western slopes of the Sierra Nevada 500 km to the south
and suggests temperatures 5°C warmer than at present and
significant summer drought.
Middle Miocene through Pliocene History 261
Figure 7.9. Pollen percentage diagram il-
lustrating paleoenvironmental fluctua-
tions at Tulelake, California from 3 to 2
Ma. Reprinted from Thompson (1991;
based on Adam et al., 1990b) with the per-
mission of Quaternary Science Reviews,
Pergamon Press, and Elsevier Science Ltd.
The Deschutes flora of northern Oregon (-4-5 Ma;
Chaney, 1938) is a low-diversity, riparian vegetation of de-
ciduous hardwoods bordered by shrubland [Acer negun-
doides, Populus washoensis, P. alexanderi (P. aff. tri-
chocarpa), Prunus irvingii (P. aff. emarginata), Quercus
prelobata, Salix sp. (S. cf. florissantii), Salix aff. caudata,
Ulmus afftnis]. The flora reflects a decline in summer pre-
cipitation from earlier Miocene times. The Palouse Prairie
of Idaho, eastern Washington, and eastern Oregon appears
in the Late Pliocene ~3 Ma, fully 10 Ma after the Plains
grassland. Its history is associated with the rise of the Cas-
cade Mountains and the development of different rainfall
regimes east and west of the Rocky Mountains. To the east,
summer rains are brought into the Plains from tropical lows
moving northward across the region in June and July. To the
west, the Pacific summer air is dry from the rainshadow ef-
fect of the Cascade Mountains and precipitation comes
from winter lows. In general, the Columbia Plateau floras
indicate that components of grassland and near-desert veg-
etation became important during the Pliocene (Leopold and
Wright, 1985) but were not widespread until the Plio-Pleis-
tocene. This is in contrast to an earlier development sug-
gested by grazing mammals (see Chapter 3, Faunas). In the
Pacific Northwest winter and summer average temperatures
in the Neogene differed by ~20°C (Wolfe, 1978).
The effects of the temperature and rainfall decline with
greater seasonality at ~15 Ma on western North American
vegetation is recorded in numerous tectonic basins where
sediments of Tertiary age preserve an extensive sequence of
floras (Axelrod, 1992b). These floras can be arranged in
262 Late Cretaceous and Cenozoic History of North American Vegetation
pairs, one just before and one just after the temperature de-
cline at -15 Ma, and along a line from south (Nevada) to
north (Washington). The Fingerrock flora (16.4 Ma; 15.5 Ma
Wolfe et al., 1997) of southwestern Nevada has numerous
deciduous hardwoods that include exotics from eastern
North America and eastern Asia (48%; species of Acer,
Alnus, Betula, Carya, Diospyros, Eugenia, Gymnocladus,
Malus, Photinia, Populus, Quercus, Robinia, Sophora,
Ulmus, and Zelkova), along with Abies, Picea, Pinus, and
Tsuga from the highlands and Arbutus, Cercocarpus, Gar-
rya, Lithocarpus, and Quercus from a drier broad-leaved
evergreen sclerophyll woodland. The Stewart Valley flora
from the same region is -14.5 m.y. in age and contains
none of the eastern exotic hardwoods. Rather, it is a
Madrean vegetation typical of present-day Arizona, south-
ern California, and especially northern Mexico where
many dry-habitat species of Pinus and Quercus likely had
their origin. The fossil flora consists of Arbutus, Garrya,
Juglans, Lyonothamnus, Populus, Quercus, Rhus, and
Sapindus, with Abies, Chamaecyparis, Pinus, Amelan-
chier, Arctostaphylos, Holodiscus, Philadelphus, Prunus,
and Ribes on the cooler slopes and Salix along river mar-
gins. A similar pattern is seen in assemblages at Pyramid
(15.6 Ma; 44% exotic species) and Purple Mountain (14.7
Ma; 14% exotics) in west-central Nevada. Further north in
northwestern Nevada the 49-Camp flora (-16 Ma) has 55%
exotics among species of Acer, Ailanthus, Carya, Cedrela,
Cercidiphyllum, Cocculus, Fagus, Nyssa, Ptelea, Ore-
opanax, Populus, Quercus, Sassafras, Tilia, and Ulmus,
with an upland flora of Abies, Chamaecyparis, Pinus, Acer,
Fraxinus, Populus, Prunus, and Rosa, and a meager broad-
leaved, evergreen sclerophyll vegetation of Arbutus and
Quercus. The Gillam Spring flora has 26% exotics. A de-
crease in summer precipitation, estimated at 35-40%
below the pre-15 Ma annual total, is inferred from the
Gillam Spring assemblage (Axelrod and Schorn, 1994). In
central Oregon floras at Mascall (-16 Ma; 70% exotic
species) and Stinking Water (11 Ma; 35%: Fig. 7.10) show a
similar trend, as do the Latah (15.8 Ma; 75% exotics) and
Ellensburg (14 Ma; 44% exotics) floras of Washington.
These data indicate that the assemblages older than -15
Ma are richer in mesic, Asian exotics and that the percent-
ages increase to the north (Mascall and Latah floras are
richest) where precipitation was higher. Floras in the same
region that are younger than -15 m.y. have fewer exotics
that were rapidly eliminated beginning in the south as
summer rainfall decreased. Modern rainfall records and
the distribution of woody taxa in the western Great Plains
suggest that when precipitation falls below 75-90 mm
(3.0-3.5 in.) during each of the summer months, decidu-
ous hardwoods rapidly decrease (Axelrod, 1992b). This is
probably near the threshhold level reached in the arid
southwest after -15 Ma. The cause for decreasing rainfall
in the southwest is complex and not all factors have been
integrated into a fully satisfactory model. The 18O/16O ma-
rine record shows a sharp decrease in benthic tempera-
tures at -15 Ma associated with the spread of the East
Antarctic ice sheet. This allowed cold waters to spread into
the world's ocean basins, which caused an upwelling of
nutrient-rich cold waters along some coastal regions. That
such an upwelling did occur is documented by widespread
blooms of diatoms preserved as extensive deposits of di-
atomite at about the same time in California and other parts
of the world. Winds blowing across these cooler waters lost
moisture and provided less rainfall to the interior. The
cooling trend was augmented at -8-7 Ma by the spread of
the West Antarctic ice sheet (intensification of high-pressure
systems under the descending arm of the Hadley convec-
tion cell), early glaciation in the northern Coast Ranges and
the Arctic Basin, and continued rise of the western moun-
tains. Axelrod (1992b) identifies the Mio-Pliocene as the
driest part of the Tertiary, and woodland-shrubland/chap-
arral spread more widely in the west and exotics persisted
only as relicts along the mild coast. Thompson (1991,
1996) provides a summary of general climatic conditions
inferred from megafossil and microfossil floras and associ-
ated faunas from the western United States: 4.8-2.4 Ma,
mostly wetter and warmer than at present; 2.4-2.0 Ma,
colder and drier; 2.0-1.8 Ma, return to warmer and more
moist.
One consequence of the Late Tertiary increase in the
height of the Rocky Mountains and other western mountain
systems was the blockage of warm air into the western Arc-
tic region of North America. Prior to the Alaska Range at-
taining significant heights at -4-5 Ma, southern Alaska
was relatively flat. The subsequent uplift contributed to the
Neogene cooling in the high latitudes. The Seldovia Point
flora of Alaska (late Early to early Middle Miocene; Chapter
6) is a deciduous forest, and a coniferous forest is in the
vicinity (Wolfe and Tanai, 1980). Although the data are
meager, the aggregation of Abies, Larix, Picea, Tsuga, and
others appears to have taken place there beginning in the
Late Miocene and Pliocene. Because several of these ele-
ments, along with Betula, are present in the Middle Eocene
Republic flora of northeastern Washington, the coniferous
forest probably spread north and south and into lower ele-
vations from upland centers in the northern Rocky Moun-
tains (Axelrod et al., 1991). In floras of Mio-Pliocene age
from the Alaska Range, Abies, Larix, Picea, Pinus, Tsuga,
Populus, and Salix were present; in the lowlands of the
Cook Inlet region of Alaska, some boreal conifers were pres-
ent by the Pliocene (Picea, Tsuga; Wolfe, 1966, 1972; Wolfe
and Hopkins, 1967; Wolfe and Leopold, 1967; Wolfe et al.,
1966). However, the Cook Inlet flora also includes some
holdovers from older Miocene vegetation, such as Glyp-
tostrobus, and possibly Pterocarya, Tilia, and Ulmus (ex-
tinct in Alaska in the Late Miocene). Wolfe et al. (1966) pro-
pose three paleobotanical stages for the Seldovia Point
Kenai Group of the Cook Inlet region: Seldovian (deciduous
forest), Homerian (mixed conifer and hardwood forest), and
Clamgulchian (conifer forest with occasional hardwoods).
The floras are similar to the modern northern mixed

conifer-hardwood forest of New England and the lake
states, with some added Asian elements, and grew under an
estimated MAT of ~9°C decreasing to ~3°C toward the top
of the sequence. [See papers in Thompson, 1994; fossil
spruce and larches from near Circle, east-central Alaska,
were studied by Miller and Ping (1994); and larches from
other parts of the Arctic are under study by Schorn (1994).]
At Fort Yukon the present MAT is -6.4°C and annual pre-
cipitation is 168 mm. A fossil palynoflora from the Porcu-
pine River region of east-central Alaska (67° 20' N) was
dated at 15.2 Ma (Seldovian) by the 40Ar/39Ar method
(White and Ager, 1994). The vegetation is deciduous forest
and includes TCT, Abies, Larix-Pseudotsuga, Picea, Pinus,
Sciadopitys (presently Asian), Tsuga, cf. Juncus, AcerC?},
Alnus, Betula, Cornus, cf. Corylus, cf. Carpinus, Carya, Cas-
tanea-type, Cercidiphyllum, Fagus, cf. Galium, Ilex,
Juglans, Liquidambar, Ludwigia, Nymphaea, Nyssa, Ptero-
carya (Asian), Quercus, Rhus types, Salix, Tilia-type, and
Wmus-type. Another pollen-spore sequence from the Ne-
nana coal field of Alaska (63.5° N) correlates with the Upper
Seldovian stage (late-early to early Middle Miocene) and is
also mixed northern hardwood forest with eastern United
States-Asian affinities and thermophilous taxa (e.g., Fagus,
Quercus}. The Middle Miocene MAT is estimated at ~9°C,
Middle Miocene through Pliocene History 263
Figure 7.10. (A) Quercus dayana,
Stinking Water. (B) Populus lind-
greni, Stinking Water. (C) Alnus har-
neyana, Stinking Water. (D) Quercus
prelobata, Stinking Water. Reprinted
from Chaney and Axelrod (1959)
with the permission of the Carnegie
Institution of Washington.
falling to ~5°C in the Late Miocene (Leopold and Liu, 1994).
An early Middle Pliocene pollen-spore assemblage at Cir-
cle has increased boreal conifers and suggests a MAT of
-0.6-2.6°C, or ~7-9°C warmer than the present MAT of
-6.4°C (Ager et al., 1994). Paleosol evidence is consistent
with deciduous forest vegetation and a temperate climate
(Smith et al., 1994). By the Late Miocene thermophilous
genera such as Carya, Fagus, Liquidambar, and Nyssa are
rare or absent. The Pliocene Lava Camp flora on the Seward
Peninsula, about 75 km south of the Arctic Circle, includes
Picea glauca, P. mariana, P. sitchensis, Pinus monticola,
Tsuga heterophylla, Carex, Cyperus, Alnus, Betula, Corylus,
Epilobium, Ericaceae, Oenothera, Salix, Symphoricarpos,
and Vaccinium (Hopkins et al., 1971). Fossil woods in the
early Middle Miocene from the Porcupine River region are
Taxodiaceae and pines with growth rings greater than 1 cm
wide, while in the Late Pliocene the woods are Picea and
Larix with growth rings less than 1 mm wide (Wheeler and
Arnette, 1994). Axelrod et al. (1991) note that this is not a
taiga community and compare it to the modern vegetation
along the northern border of the coast coniferous forest. The
associated insect fauna is not a tundra or boreal forest
assemblage but is similar to that occurring in the Pacific
coast coniferous forest of southern British Columbia-
264 Late Cretaceous and Cenozoic History of North American Vegetation

Table 7.9. Pollen and spore counts of Colvillian age, North Slope of Alaska.

Site 1 Sample Site 2 Sample

A B C A B

Abies — + + +
—
Tsuga mertensiana — + + — —
Other Tsuga — + + — +
Pinus 2.3 8.8 8.5 10.2 1.2
Picea 25.1 30.5 18.7 33.8 13.9
Alnus 5.1 6.6 5.7 7.5 16.3
Betula 36.0 27.4 33.5 27.5 32.2
Ericales 5.1 6.0 8.2 6.9 6.6
Salix 1.6 1.6 2.2 — +
Populus 1.3 2.5 + 1.3 +
Other woody taxa Co Ca,La Co — Co
Gramineae 7.7 4.4 8.9 4.6 14.8
Cyperaceae 10.3 6.6 10.4 3.9 9.9
Other herbs 3.2 1.9 1.6 2.6 +
Unknown + 1.9 + + 1.2
Indeterminate 9.6 7.2 8.5 13.8 9.3
Sphagnum 10.0 14.5 19.6 17.4 14.5
Lycopodium selago — + + + +
Other Lycopodium 6.8 4.7 2.2 7.8 4.5
Selaginella selaginoides + — + 1.3 +
Other Selaginella + — — + —
Cf. Osmunda 2.6 1.9 + — +
Other trilete spores 19.0 11.9 23.4 8.8 11.1
Monolete fern spores 3.9 2.2 7.3 5.6 1.2
Pediastrum colonies + — + — 2.7

All values are expressed as a percent of the total pollen excluding indeterminate grains. (—) Not present; (+) present but <1%; Co, Corylus; Ca, Carya;
La, Larix. Adapted from Nelson and Carter (1992).

northern Washington. This forest extended to near the Arc-
tic Circle in the Pliocene, as shown by the Kivalina flora
100 km north of Lava Camp (Hopkins et al., 1971). A boreal
forest of modern aspect was not yet assembled or wide-
spread. Rather, vegetation in the Arctic began changing
from a mixed mesophytic hardwood forest to a coniferous
forest at -15 Ma followed by a more prominent herb flora at
-7 Ma (White et al., 1997). By the end of the Tertiary aver-
age July temperatures in the Cook Inlet region dropped by
~7°C from the Early Miocene thermal high (Wolfe and
Leopold, 1967; Wolfe and Tanai, 1980). It was not until ~4
Ma that the lowland boreal coniferous forest became estab-
lished.
On the North Slope of Alaska on the Arctic Ocean bor-
derland (~70° N), a series of coastal plain sediments in the
Gubik Formation were deposited during the Late Pliocene
and Pleistocene. Several transgressions are evident; the
late Pliocene ones are called the Colvillian (oldest), Big-
bendian, and Fishcreekian. The time interval from the
Colvillian through the Bigbendian is from ~3 to -2.4 Ma
when the latest Bigbendian was being deposited near the
beginning of the Matuyama Reversed polarity chron. The
end of the Fishcreekian is placed at -2 Ma. Ostracodes
from the Colvillian and Bigbendian indicate that waters
were still warmer than at present and comparable to those
today along the coasts of northern Nova Scotia and south-
ern Labrador (45°-50° N; viz., not frozen to the shore any
time during the year; Repenning and Brouwers, 1992). In
the Fishcreekian the ostracode fauna suggests cooling to
~5°C, similar to temperatures now prevailing in the south-
ern Bering Sea (60°-65° N).
Pollen from the Ocean Point section on the Colville
River (Table 7.9; Fig. 7.11) also indicates a climate warmer
than at present (Nelson and Carter, 1985, 1992). Tsuga and
Pinus were present in the Colvillian but decreased or dis-
appeared in the Bigbendian, leaving a spruce-birch park-
land with some Pinus, Tsuga, Larix, Alnus, and Populus.
There was no extensive tundra or permafrost as there is
today, but open areas did include grasses and tundra ele-
ments (forest-tundra mosaic). In the late Fishcreekian (~2
Ma) the boreal forest still extended further north than at
present, as suggested by occasional pollen of Larix, Picea,
and Pinus; but the local vegetation was changing through a
shrub tundra to herb-dominated tundra, and permafrost
was present.
A summary of the age of these western North American
floras between 16 and 2 Ma, along with proposed floral
stages (Wolfe, 1981), is given in Fig. 7.12.
Figure 7.11. Pollen percentage diagram from the Gubik Formation at the Ocean Point site. Note changes in scale. 'Other
taxa: a, Populus; b, Menyanthes; c, Potamogeton; d, Plantago; e, Listera type (Orchidaceae); f, Drosera; g, Myriophyllum;
and h, Rubiaceae. Reprinted from Nelson and Carter (1985) with the permission of Quaternary Research.

Figure 7.12. Summary of the relative ages, floral stages, and

radiometric dates for the principal western North Ameri-
can floras between -16 and 2 m.y. in age discussed in the
text.
266 Late Cretaceous and Cenozoic History of North American Vegetation
VEGETATION SUMMARY
Although published information on vegetational history is
surprisingly meager for the Middle Miocene through the
Pliocene of the southeastern United States, it is likely that
modernization of the vegetation accelerated with the Mid-
dle to Late Miocene cooling event (Table 7.10). The princi-
pal effects were the disappearance of most Asian and
neotropical exotics by the Pliocene, modernization and ex-
pansion of the mixed mesophytic forest, the temporal asso-
ciation of different forest genera to constitute various other
deciduous forest associations, and development of the Ap-
palachian coniferous forest. In Chapter 6 it was noted that
by Claiborne (Middle Eocene)-Jackson (Late Eocene) time,
elements were available for assemblage into the mixed meso-
phytic, southern mixed hardwood, oak-hickory, oak-
chestnut, sand pine scrub (Pinus, Sabal, other rosette
palms), flood-plain (Taxodium, Nyssa), and mangrove
(Acrostichum, Nipa) associations. Tilia is reported from
the Claiborne Formation, but AcerC?) first occurs in the
Jackson Formation. Definite Acer pollen is known from the
Brandywine flora, so an initial maple-basswood associa-
tion in the region could date from the Late Miocene. Nipa
disappeared by the end of the Eocene, but it is not known
when Rhizophora first appeared in the southeast. In the in-
tervening period, brackish habitats were probably occu-
pied by Acrostichum and various palms. Picea is known
from the Jackson Formation, Tsuga occurs later in the Mid-
dle Miocene Legler Lignite, and Abies is found in Pliocene
deposits but to the north at Martha's Vineyard. Thus, the
limited fir-spruce-hemlock Appalachian coniferous for-
est in the south is likely of Plio-Pleistocene origin. It may
have formed earlier to the north, but Abies and Picea are
not known from the Early Miocene Brandon Lignite of Ver-
mont. Today Abies balsamea (balsam fir) extends south on
isolated peaks to Virginia and West Virginia, and A. fraseri
(Fraser fir) reaches only as far south as North Carolina and
Tennessee at elevations of 1400-2100 m (4500-6900 ft).
To the north, the deciduous forest intermingled with an
increasing number of cold-temperate conifers to form a
version of the lake states forest. This community was more
broadly distributed than at present; however, with later de-
velopment of the boreal coniferous forest, it gradually be-
came limited and better defined as the transition vegeta-
tion between the deciduous forest and boreal coniferous
forest in the Great Lakes region. Further to the north, fossil
floras record a mixed temperate hardwood forest (Middle
Miocene), a cool-temperate coniferous-hardwood forest
(Mio-Pliocene), and a cold-temperate conifer forest and
Alnus woodland (late Pliocene). In the Arctic islands the
sequence was from conifer forest, to stunted conifer wood-
land, to depauperate forest-tundra vegetation (Plio-
Pleistocene, -1.7 Ma).
Boreal coniferous forest elements such as Abies, Picea,
Thuja, Tsuga, and Betula are first recorded together in the
Middle Eocene Republic flora of northwestern Washington;
these volcanic highlands were likely a center of dispersal
for the formation. The forest was moving into lower eleva-
tions by -15 Ma, and by the Mio-Pliocene a version was
occupying the lowlands of the Cook Inlet region of Alaska;
but it still included some Asian exotics and deciduous
hardwoods (Glyptostrobus, Pterocarya, Tilia, Ulmus). These
disappeared in the Pliocene, and the Lava Camp and Ki-
valina floras preserve a near-taiga community most similar
to the modern coast coniferous forest of northern Washing-
ton and southern British Columbia. The boreal coniferous
forest of essentially modern aspect appears later in the
Pliocene at -4 Ma, and by 2 Ma it had reached the Colville
River region on the north slope of Alaska. With the contin-
uing trend toward colder climates it began disappearing
along its northern limits at this time and was locally mixed
with shrub tundra and herb-dominated tundra growing on
permafrost. Widespread tundra of modern aspect did not
develop until near the onset of extensive Arctic continental
glaciations at -850 ka.
In the continental interior, the trend toward colder and
winter-dry climates had produced a mosaic of open decid-
uous forest-woodland with patches of grassland by -13
Ma (Kilgore flora). Between 13 and 6-5 Ma grassland ex-
panded at the expense of forest that was confined mostly to
riparian habitats. Near-prairie vegetation with Juniperus
and Celtis occupied much of the region. Grasslands of
modern aspect appeared with the next principal tempera-
ture decline beginning -1.6 Ma and later in the Quaternary.
The woodland-grassland of the Plains graded eastward
into a pifion-juniper woodland toward the eastern slopes
of the central and northern Rocky Mountains. The high-
lands supported a western montane coniferous forest. The
woodland extended around the southern end of the Rocky
Mountains into the southwest where it was associated with
evergreen oak to form Madrean woodland-chaparral. By
the Pliocene the southwest was arid and the vegetation re-
sembled the more mesic parts of the Sonoran Desert. Along
the moister southern California coast the vegetation was a
pine—oak woodland that included palms and Persea
(Chula Vista flora). To the north in the San Francisco Bay
area (Sonoma flora) the lowland vegetation near the coast
was a Pacific coast coniferous forest that included Sequoia,
while further inland a pine-douglas fir forest grew in
moister habitats with an oak woodland-chaparral on the
slopes.
The principal Miocene communities of the Columbia
Plateau region during the Miocene (Succor Creek flora)
were a mixed mesophytic association of the deciduous for-
est generally below -600 m that included Asian and east-
ern North American exotics, a transition conifer-hard-
wood forest, and an upland western montane coniferous
forest. The conifer-hardwood forest extended northward
into Alaska (Seldovia Point flora) and downward in eleva-
tion during the Middle Tertiary. The decline of the mixed
mesophytic association in western North America is
recorded in the Idaho Group floras with the rise of the
Table 7.10. Summary of North American vegetation types and estimated MATs, Middle Miocene through Pliocene.
Region Vegetation MAT

Middle to Late Miocene

Southeast Warm-temperate deciduous
Mid-Atlantic Warm-temperate deciduous
New Jersey (mixed mesophytic, beech-maple,
Kirkwood maple-basswood, lake states,
Legler (11 Ma) oak-chestnut, oak-hickory,
Maryland southern mixed hardwood, sand
Brandywine pine scrub, floodplain); Appalachian montane
coniferous (north), early version (south)
Northeast
Massachusetts
Martha's Vineyard Warm-temperate deciduous
Iceland
Holmatindur (10.3-9.5 Ma) Sequence: deciduous to subarctic Alnus woodland

Western interior
Nebraska
Kilgore(13-14Ma) Woodland-grassland, riparian deciduous
Ash Hollow (6-7 Ma) Wooded grassland

West
Idaho
Trapper Creek (10.5-12 Ma) Deciduous, western montane coniferous
Idaho Group (Miocene part) Deciduous, pine woodland
Wyoming
Teewinot (8 Ma) Shrubland to near desert, riparian, western
montane coniferous
Nevada
Fingerrock (16.4) Ma), Pyramid (15.6 Ma) Deciduous 13.5°C
Verdi Oak woodland, chaparral, savanna
Stewart Valley (14 Ma) Pine-oak woodland (Madrean)
California
Mount Eden (5.3 Ma) Near desert

Northwest
Oregon
Succor Creek (14-15 Ma) Deciduous, swamp cypress (floodplain), riparian,
mixed conifer-hardwood, western montane coniferous
Washington
Ellensburg Swamp cypress (floodplain), deciduous, oak
woodland (drier sites)
Alaska
Seldovia, Cook Inlet, Lava Camp, Kivalina Deciduous, mixed conifer-hardwood, 9-3°C
Pacific coast coniferous, early boreal coniferous
Pliocene
Northeast
Massachusetts
Martha's Vineyard Cold-temperate deciduous
Greenland
Kap K0benhavn (82° 30' N; 2 Ma) Boreal forest, forest tundra

West
Idaho
Idaho Group (Pliocene part) Desert-shrubland with grasses
California
Chula Vista (3 Ma) Pine-oak woodland, riparian warm temperate 16°C
Sonoma
Napa Western montane coniferous, oak-woodland-chaparral
Santa Rosa Pacific coast coniferous (Sequoia)
Tulelake(3-2Ma) Western montane coniferous

Alaska
Colville Boreal coniferous, forest-tundra
Plant formation and association names follow terminology used for modern communities discussed in Chapter 1 and listed in Table 1.1.
268 Late Cretaceous and Cenozoic History of North American Vegetation
coastal mountains. Impoverished pine forest, then Great
Basin desert-shrubland (Sarcobatus, Artemisia] with
grasses appear at -3-2 Ma.
Important events in the modernization of the North
American vegetation during the period from -16.3-1.6 Ma
were:
1. the disappearance of Asian and neotropical exotics
by the Pliocene (-5 Ma);
2. the continued but temporal grouping of various gen-
era into associations that define the present version
of the deciduous forest formation (e.g., maple-bass-
wood);
3. the persistence of tropical elements along the
Florida coast;
4. the further association of Abies, Picea, Tsuga, and
others to constitute the Appalachian montane forest
(progressively later toward the south);
5. the appearance of a near-modern boreal coniferous
forest at -4 Ma;
6. the gradual restriction of the transition vegetation
between the boreal coniferous forest and the decid-
uous forest into the present lake states forest;
7. the development of near-grasslands in the continen-
tal interior by -2 Ma;
8. the formation of near-modern deserts during the
Pliocene (-5 Ma) with the continued rise of the
Sierra Nevada, initial appearance of the high Cas-
cade Mountains and the Coast Ranges, and redirec-
tion of atmospheric circulation;
9. the appearance of the high-altitude coastal conifer-
ous forest in the midlatitudes by -4 Ma;
10. the consequent disruption of the earlier near-
continuous deciduous forest formation across the
temperate latitudes of the northern hemisphere;
11. the appearance of the California grasslands and
Palouse Prairie at -3 Ma; and
12. the appearance of near-modern tundra and sus-
tained permafrost at -2 Ma.
Beginning at -1.6 Ma, these Late Tertiary communities
were subjected to the increasingly turbulent and fast-paced
climatic events of the oncoming ice age.
References
Adam, D. P. et al. (+ 5 authors). 1989. Tulelake, California:
the last 3 million years. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 72: 89-103.
J. R Bradbury, and H. J. Rieck. 1990a. Pliocene environmen-
tal changes at Tulelake, Siskiyou County, California,
from 3 to 2 Ma. In: L. B. Gosnell and R. Z. Poore (eds.),
Pliocene Climates: Scenario for Global Warming. U.S.
Geological Survey, Reston, VA. Open-File Report
90-64, pp. 3-5.
and A. M. Sarna-Wojcicki. 1990b. Environmental
changes in the Tule Lake Basin, Siskiyou and Modoc
Counties, California, from 3 to 2 million years before
present. U.S. Geol. Surv. Bull. 1933:1-13.
Ager, T. A., J. V. Matthews, Jr., and W. Yeend. 1994. Pliocene
terrace gravels of the ancestral Yukon River near Cir-
cle, Alaska: palynology, paleobotany, paleoenviron-
mental reconstruction and regional correlation.
Quatern. Int. 22/23: 185-206.
Armstrong, R. L., W. P. Leeman, and H. E. Malde. 1975. K-
Ar dating, Quaternary and Neogene volcanic rocks of
the Snake River Plain, Idaho. Am. J. Sci. 275: 225-251.
Arroyo, M. T. K., P. H. Zedler, and M. D. Fox (eds.). 1995.
Ecology and Biogeography of Mediterranean Ecosys-
tems in Chile, California, and Australia. Ecological
Studies 108. Springer-Verlag, Berlin.
Austin, J. J., A. J. Ross, A. B. Smith, R. A. Fortey, and R. H.
Thomas. 1997. Problems of reproducibility—does ge-
ologically ancient DNA survive in amber-preserved in-
sects? Proc. Roy. Soc. London B . 264: 467-474.
Axelrod, D. I. 1937. A Pliocene flora from the Mount Eden
beds, southern California. Carnegie Inst. Wash. Con-
trib. Paleontol. 476: 125-183.
. 1940. A record of Lyonothamnus in Death Valley,
California. J. Geol. 48: 526-531.
. 1944. The Sonoma flora (California). Carnegie Inst.
Wash. Contrib. Paleontol. 533: 167-206.
. 1950a. The Anaverde flora of southern California.
Carnegie Inst. Wash. Contrib. Paleontol. 590:119-158.
. 1950b. The Pirn Gorge flora of southern California.
Carnegie Inst. Wash. Contrib. Paleontol. 590:159-214.
. 1950c. Further studies of the Mount Eden flora,
southern California. Carnegie Inst. Wash. Contrib. Pa-
leontol. 590: 73-117.
. 1950d. A Sonoma florule from Napa, California.
Carnegie Inst. Wash. Contrib. Paleontol. 590: 23-71.
. 1956. Mio-Pliocene floras from west-central
Nevada. Univ. Calif. Publ. Geol. Sci. 33: 1-322.
. 1958. The Pliocene Verdi flora of western Nevada.
Univ. Calif. Publ. Geol. Sci. 34: 91-160.
. 1964. The Miocene Trapper Creek flora of southern
Idaho. Univ. Calif. Publ. Geol. Sci. 51: 1-180.
. 1978. Origin of coastal sage vegetation, Alta and
Baja California. Am. J. Bot. 65: 1117-1131.
. 1980. Contributions to the Neogene paleobotany of
central California. Univ. Calif. Publ. Geol. Sci. 121:
1-212.
. 1985. Rise of the grassland biome, central North
America. Bot. Rev. 51: 163-201.
. 1987. The Late Oligcene Creede flora, Colorado.
Univ. Calif. Publ. Geol. Sci. 130: 1-235.
. 1992a. The Middle Miocene Pyramid flora of west-
ern Nevada. Univ. Calif. Publ. Geol. Sci. 137:1-50.
. 1992b. Miocene floristic change at 15 Ma, Nevada
to Washington, U.S.A. Palaeobotanist 41: 234-239.
. 1995. The Miocene Purple Mountain flora of west-
ern Nevada. Univ. Calif. Publ. Geol. Sci. 139:1-62.
and J. Cota. 1993. A further contribution to the
closed-cone pine (Oocarpae) history. Am. J. Bot. 80:
743-751.
Axelrod, D. I. and T. A. Demere. 1984. A Pliocene flora
from Chula Vista, San Diego County, California. Trans.
San Diego Soc. Nat. Hist. 20: 277-300.
Axelrod, D. I., M. T. Kalin Arroyo, and P. H. Raven. 1991.
Historical development of temperate vegetation in the
Americas. Rev. Chil. Hist. Nat. 64: 413-446.
Axelrod, D. I. and H. E. Schorn. 1994. The 15 Ma floristic
crisis at Gillam Spring, Washoe County, northwestern
Nevada. PaleoBios 16:1-10.
Barnowsky C. W. 1984. Late Miocene vegetational and cli-

matic variations inferred from a pollen record in
northwest Wyoming. Science 223: 49-51.
Barrett, E }., C. J. Adams, W. C. Mclntosh, C. C. Swisher III,
and G. S. Wilson. 1992. Geochronological evidence
supporting Antarctic deglaciation three million years
ago. Nature 359: 816-818.
Bennike, O. and J. Bocher. 1990. Forest-tundra neighbour-
ing the North Pole: plant and insect remains from the
Plio-Pleistocene Kap K0benhavn Formation, north
Greenland. Arctic 43: 331-338.
Berggren, W. A., D. V. Kent, M.-P. Aubry, and J. Hardenbol
(eds.). 1995. Geochronology, Time Scales and Global
Stratigraphic Correlations. Society for Sedimentary
Geology, Tulsa, OK. Special Publication 54.
Brooks, B. W. 1935. Fossil plants from Sucker Creek, Idaho.
Ann. Carnegie Mus. 24: 275-337.
Call, V. B. and D. L. Dilcher. 1995. Fossil Ptelea samaras
(Rutaceae) in North America. Am. J. Bot. 82:
1069-1073.
Cano, R. J., H. N. Poinar, D. W Roubik, and G. O. Poinar, Jr.
1992. Enzymatic amplification and nucleotide se-
quencing of portions of the 18S rDNA gene of the bee
Prolebeia dominicana (Apidae: Hymenoptera) isolated
from 25-40 million year old Dominican amber. Med.
Sci. Res. 20: 619-622.
Chamley, H. 1979. North Atlantic clay sedimentation and
paleoenvironment since the Late Jurassic. In: M. Tal-
wani, W. Hay, and W. B. E Ryan (eds.), Deep Drilling
Results in the Atlantic Ocean: Continental Margins
and Paleoenvironment. Amer. Geophysical Union,
Washington, D.C. pp. 342-361.
Cerling, T. E. et al. (+ 6 authors). 1997. Global vegetation
change through the Miocene/Pliocene boundary. Na-
ture 389: 153-158.
Chaney, R. W. 1938. The Deschutes flora of eastern Oregon.
Carnegie Inst. Wash. Contrib. Paleontol. 476:185-216.
. 1959. Miocene floras of the Columbia Plateau. Part
I. Composition and interpretation. Carnegie Inst.
Wash. Contrib. Paleontol. 617: 1-134.
and D. I. Axelrod. 1959. Miocene floras of the Co-
lumbia Plateau. Part II. Systematic considerations.
Carnegie Inst. Wash. Contrib. Paleontol. 617:135-237.
Chaney, R. W. and M. K. Elias. 1938. Late Tertiary floras
from the High Plains. Carnegie Inst. Wash. Contrib. Pa-
leontol. 476: 1-72.
Coates, A. G. et al. (+ 7 authors). 1992. Closure of the Isth-
mus of Panama: the near-shore marine record of Costa
Rica and western Panama. Geol. Soc. Am. Bull. 104:
814-828.
Condit, C. 1938. The San Pablo flora of west central Cali-
fornia. Carnegie Inst. Wash. Contrib. Paleontol. 476:
217-268.
Cronin, T. M. 1991. Pliocene shallow water paleoceanogra-
phy of the North Atlantic Ocean based on marine os-
tracodes. Quat. Sci. Rev. 10: 175-188.
and H. J. Dowsett (eds.). 1991. Pliocene Climates.
Quat. Sci. Rev. 10: 1-296 [special issue].
Cross, A. T. and R. E. Taggart. 1982. Causes of short-term
sequential changes in fossil plant assemblages: some
considerations based on a Miocene flora of the north-
west United States. Ann. Missouri Bot. Garden 69:
676-734.
Crowley, T. J. 1991. Modeling Pliocene warmth. Quat. Sci.
Rev. 10: 275-282.
Middle Miocene through Pliocene History 269
and G. R. North. 1991. Paleoclimatology. Oxford
Monographs in Geology and Geophysics.Oxford Uni-
versity Press, Oxford, UK. No. 16.
Curry, W. B. and K. G. Miller. 1989. Oxygen and carbon iso-
topic variation in Pliocene benthic foraminifers of the
equatorial Atlantic. Ocean Drill. Program Sci. Results
108: 157-166.
Davis, O. K. In press. Preliminary pollen analysis of Ceno-
zoic sediments of the Great Salt Lake, U.S.A. Ameican.
Association of Stratigraphic Palynologists Pliocene
Symposium, Baton Rouge, LA, October 1993.
DeSalle, R., J. Gatesy, W. Wheeler, and D. Grimaldi. 1992.
DNA sequences from a fossil termite in Oligo-
Miocene amber and their phylogenetic implications.
Science 257: 1933-1936.
Dorf, E. 1930. Pliocene floras of California. Carnegie Inst.
Wash. Contrib. Paleontol. 412: 1-112.
Downing, K. E and C. C. Swisher III. 1993. New 40Ar/39Ar
dates and refined geochonology of the Sucker Creek
Formation, Oregon. J. Vert. Paleontol. 13: 3 3A.
Dowsett, H. J. and T. M. Cronin. 1990. High eustatic sea
level during the Middle Pliocene: evidence from the
southeastern U.S. Atlantic Coastal Plain. Geology 18:
435-438.
Dowsett, H. J. and P. Loubere. 1992. High resolution Late
Pliocene sea-surface temperature record from the
northeast Atlantic Ocean. Mar. Micropaleontol. 20:
91-105.
Elias, M. K. 1932. Grasses and other plants from the Ter-
tiary rocks of Kansas and Colorado. Univ. Kansas Sci.
Bull. 20: 333-367.
. 1935. Tertiary grasses and other prairie vegetation
from the High Plains of North America. Am. J. Sci.,
Ser. 5, 29: 24-33.
-. 1942. Tertiary Prairie Grasses and Other Herbs from
the High Plains. Geology Society of America, Boulder,
CO. Special Paper 41.
Emslie, S. D. and G. S. Morgan. 1994. A catastrophic death
assemblage and paleoclimatic implications of Pliocene
seabirds of Florida. Science 264: 684-685.
Essay Collection. 1996. Marine Geology 27.
Evernden, J. E and G. T. James. 1964. Potassium-argon dates
and the Tertiary floras of North america. Am. J. Sci.
262: 945-974.
Eyde, R. H. and E. S. Barghoorn. 1963. Morphological and
paleobotanical studies of the Nyssaceae, II the fossil
record. J. Arnold Arb. 44: 328-370.
Fields, P E 1983. A Review of the Miocene Stratigraphy of
Southwestern Idaho, with Emphasis on the Payette
Formation and Associated Floras. Masters thesis. Uni-
versity of California, Berkeley, CA.
. 1990. Succor Creek floral taxonomic diversity: a
continuing work in progress. Maps Digest, Expo XII
Ed. 13: 50-59.
. 1992. Floras of Idaho's past: geographic and pale-
obotanic overview of the Middle Miocene Succor
Creek flora as an example. J. Idaho Acad. Sci. 28: 44-
56.
. 1996. The Succor Creek Flora of the Middle Eocene
Sucker Creek Formation, Southwestern Idaho and
Eastern Oregon: Systematics and Paleoecology. Ph.D.
dissertation. Michigan State University, East Lansing,
MI.
Frederiksen, N. O. 1984. Stratigraphic, paleoclimatic, and
270 Late Cretaceous and Cenozoic History of North American Vegetation
paleobiogeographic significance of Tertiary sporo-
morphs from Massachusetts. U.S. Geological Survey,
Reston, VA. Prof. Paper 1308, pp. 1-25.
-. 1989. Late Cretaceous and Tertiary floras, vegeta-
tion, and paleoclimates of New England. Rhodora 91:
25-48.
Funder, S., N. Abrahamsen, O. Bennike, and R. W. Feyling-
Hanssen. 1985. Forested Arctic: evidence from north
Greenland. Geology 13: 542-546.
Fyles, J. G. et al. (+ 6 authors). 1994. Ballast Brook and
Beaufort formations (Late Tertiary) on northern Banks
Island, Arctic Canada. Quatern. Int. 22/23: 141-171.
Golenberg, E. M. 1991. Amplification and analysis of
Miocene plant fossil DNA. Phil. Trans. R. Soc. Lond. B
333:419-427.
et al. (+ 6 authors). 1990. Chloroplast DNA sequence
from a Miocene Magnolia species. Nature 344: 656-658.
Graham, A. 1963. Systematic revision of the Sucker Creek
and Trout Creek Miocene floras of southeastern Ore-
gon. Am. J. Bot. 50: 921-936.
. 1965. The Sucker Creek and Trout Creek Miocene
floras of southeastern Oregon. Kent State Univ. Bull.,
Res. Ser. IX.
. 1976a. Late Cenozoic evolution of tropical lowland
vegetation in Veracruz, Mexico. Evolution 29:
723-735.
. 1976b. Studies in neotropical paleobotany. II. The
Miocene communities of Veracruz, Mexico. Ann. Mis-
souri Bot. Garden 63: 787-842.
-. 1992. Utilization of the isthmian land bridge dur-
ing the Cenozoic—paleobotanical evidence for timing,
and the selective influence of altitudes and climate.
Rev. Palaeobot. Palynol. 72:119-128.
Greller, A. M. and L. D. Rachele. 1983. Climatic limits of
exotic genera in the Legler palynoflora, Miocene, New
Jersey, U.S.A. Rev. Palaeobot. Palynol. 40:149-163.
Groot, J. J. 1991. Palynological evidence for Late Miocene,
Pliocene, and Early Pleistocene climate changes in the
middle U.S. Atlantic coastal plain. Quat. Sci. Rev. 10:
147-162.
Haggard, K. K. and B. H. Tiffney. 1997. The flora of the
Early Miocene Brandon lignite, Vermont, U.S.A. VII.
Caldesia (Alismataceae). Am. J. Bot. 84: 239-252.
Hesse, C. J. 1936. Lower Pliocene vertebrate fossils from
the Ogallala Formation (Laverne zone) of Beaver
County, Oklahoma. Carnegie Inst. Wash. Contrib. Pale-
ontol. 467: 47-72.
Hills, L. V, J. E. Klovan, and A. R. Sweet. 1974. Juglans
eocinerea n. sp., Beaufort Formation (Tertiary), south-
western Banks Island, Arctic Canada. Can. J. Bot. 52:
65-90.
Hopkins, D. M., J. V. Matthews, Jr., J. A. Wolfe, and M. L. Sil-
berman. 1971. A Pliocene flora and insect fauna from
the Bering Strait region. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 9: 211-231.
Huber, N. K. 1981. Amount and timing of Late Cenozoic
uplift and tilt of the central Sierra Nevada, Califor-
nia—evidence from the upper San Joaquin River
Basin. U.S. Geological Survey, Reston, VA. Profes-
sional Paper 1197, pp. 1-28.
Hutchison, J. H. 1982. Turtle, crocodilian, and champ-
sosaur diversity changes in the Cenozoic of the north-
central region of western United States. Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 37: 149-164.
Knowlton, E H. 1926. Flora of the Latah Formation of
Spokane, Washington, and Coeur d'Alene, Idaho. U.S.
Geological Survey, Reston, VA. Professional Paper
140A, pp. 1-81.
Krantz, D. E. 1991. A chronology of Pliocene sea-level fluc-
tuations: the U.S. Middle Atlantic Coastal Plain record.
Quatern. Sci. Rev. 10: 163-174.
Larsen, H. C. et al. (+ 6 authors). 1994. Seven million years
of glaciation in Greenland. Science 264: 952-955.
Leg 105 Shipboard Scientific Party. 1986. High-latitude
palaeoceanography. Nature 230: 17-18.
Leopold, E. B. and M. E Denton. 1987. Comparative age of
grassland and steppe east and west of the northern
Rocky Mountains. Ann. Missouri Bot. Garden 74: 841-
867.
Leopold, E. B. and G. Liu. 1994. A long pollen sequence of
Neogene age, Alaska Range. Quatern. Int. 22/23: 103-
140.
Leopold, E. B. and V C. Wright. 1985. Pollen profiles of the
Plio-Pleistocene transition in the Snake River Plain,
Idaho. In: C. J. Smiley (ed.), Late Cenozoic History of
the Pacific Northwest. Pacific Division, American As-
sociation for the Advancement of Science, San Fran-
cisco, CA. pp. 323-348.
MacGinitie, H. D. 1933. The Trout Creek flora of southeast-
ern Oregon. Carnegie Inst. Wash. Contrib. Paleontol.
416: 21-68.
. 1962. The Kilgore flora, a Late Miocene flora from
northern Nebraska. Univ. Calif. Publ. Geol. Sci. 35: 67-
158.
Manchester, S. R., P. R. Crane, and D. L. Dilcher. 1991. Nor-
denskioldia and Trochodendron (Trochodendraceae)
from the Miocene of northwestern North America. Bot.
Gaz. 152: 357-368.
Marchant, D. R., C. C. Swisher III, D. R. Lux, D. P West, Jr.,
and G. H. Denton. 1993. Pliocene paleoclimate and
East Antarctic ice-sheet history from surficial ash de-
posits. Science 260: 667-670.
Masterson, J. 1994. Stomatal size in fossil plants: evidence
for polyploidy in majority of angiosperms. Science
264:421-424.
Matthews, J. V, Jr. 1987. Plant Macrofossils from the Neo-
gene Beaufort Formation on Banks and Meighen Is-
lands, District of Franklin. Current Research, Part A,
Geologic Survey of Canada, Ottawa, Paper 87-1A, pp.
73-87.
and L. E. Ovenden. 1990. Late Tertiary plant macro-
fossils from localities in Arctic/Subarctic North Amer-
ica: a review of the data. Arctic 43: 364-392.
McCartan, L. et al. (+ 7 authors). 1990. Late Tertiary floral
assemblage from upland gravel deposits of the south-
ern Maryland coastal plain. Geology 18: 311-314.
Miller, C. N., Jr. and L. Ping. 1994. Structurally preserved
larch and spruce cones from the Pliocene of Alaska.
Quatern. Int. 22/23: 207-214.
Mudie, P. J. and J. Helgason. 1983. Palynological evidence
for Miocene climatic cooling in eastern Iceland about
9.8 Myr ago. Nature 303:689-692.
Nelson, R. E. and L. D. Carter. 1985. Pollen analysis of a
Late Pliocene and early Pleistocene section from the
Gubik Formation of Arctic Alaska. Quatern. Res. 24:
295-306.
. 1992. Preliminary interpretation of vegetation and
paleoclimate in northern Alaska during the Late
Color illustration of fossil leaf of Zelkova oregoneniana showing preservation of
flavonoid pigments for -14-15 Ma (Niklas and Giannasi, 1977; Niklas, 1981). This
was an early demonstration of the potential preservation of biochemical compounds
in megafossils of Tertiary age. Other examples of paleobiochemical fossils include pig-
ment compounds from lake sediments, alkenones (p. 32), ancient DNA (pp. 255-256),
and optically active compounds from more ancient deposits. The gradual accumula-
tion of such records may eventually make possible a more complete reading of the
Earth's biotic history in terms of molecular evolution. Photograph courtesy of Karl J.
Niklas.

Pliocene Colvillian marine transgression. U.S. Geol.
Surv. Bull. 1999: 219-222.
Niklas, K. J. 1981. The chemistry of fossil plants. Bio-
Science 31: 820-825.
and D. E. Giannasi. 1977. Flavonoides and other
chemical constituents of fossil Miocene Zelkova (Ul-
maceae). Science 196: 877-878.
Ovenden, L. 1993. Late Tertiary mosses of Ellesmere Is-
land. Rev. Palaeobot. Palynol. 79: 121-131.
Owens, J. P. et al. (+ 5 authors). 1988. Stratigraphy of the
Tertiary Sediments in a 945-Foot-Deep Corehole Near
Mays Landing in the Southeastern New Jersey Coastal
Plain. U.S. Geological Survey Professional Paper 1484,
pp. 1-39.
Pazzaglia, E J., R. A. J. Robinson, and A. Traverse. 1997. Pa-
lynology of the Bryn Mawr Formation (Miocene): in-
sights on the age and genesis of middle Atlantic mar-
gin fluvial deposits. Sedimentary Geology 108:19-44.
Poinar, H. N., M. Hoss, J. L. Bada, and S. Paabo. 1996.
Amino acid racemization and the preservation of an-
cient DNA. Science 272: 864-866.
Rachele, L. D. 1976. Palynology of the Legler lignite: a de-
posit in the Tertiary Cohansey Formation of New Jer-
sey, U.S.A. Rev. Palaeobot. Palynol. 22: 225-252.
Raymo, M. E., B. Grant, M. Horowitz, and G. H. Rau. 1996.
Mid-Pliocene warmth: stronger greenhouse and
stronger conveyor. Mar. Micropaleontol. 27: 313-326.
Raymo, M. E. and G. Rau. 1992. Plio-Pleistocene atmos-
pheric CO2 levels inferred from POM 313C at DSDP site
607. Trans. Am. Geophys. Union 73: 95-96.
. 1993. Mid-Pliocene warmth: higher CO2 or stronger
ocean heat transport? Trans. Am. Geophys. Union 74:
49.
Rea, D. K., M. Leinen, and T. R. Janecek. 1985. Geologic ap-
proach to the long-term history of atmospheric circula-
tion. Science 227: 721-725.
Renney, K. M. 1969. The Miocene Temblor Flora of West
Central California. Masters Thesis. University of Cali-
fornia, Davis, CA.
Repenning, C. A. and E. M. Brouwers. 1992. Late
Pliocene-Early Pleistocene ecologic changes in the
Arctic Ocean borderland. U.S. Geol. Surv. Bull. 2036:
1-37.
Rogers, S. O. and A. J. Bendich. 1985. Extraction of DNA
from milligram amounts of fresh, herbarium, and
mummified plant tissues. Plant Mol. Biol. 5: 69-76.
Schoell, M., S. Schouten, J. S. Sinninghe Damste, J. W. de
Leeuw, and R. E. Summons. 1994. A molecular organic
carbon isotope record of Miocene climate changes.
Science 263: 1122-1125.
Schorn, H. E. 1994. A preliminary discussion of fossil
larches (Larix, Pinaceae) from the Arctic. Quatern. Int.
22/23: 173-183.
and N. L. Gooch. 1994. Amelanchier hawkinsae sp.
nov. (Rosaceae, Maloideae] from the Middle Miocene
of Stewart Valley, Nevada, and a review of the genus in
the Nevada Neogene. PaleoBios 16: 1-17.
Smiley, C. J. 1963. The Ellensburg flora of Washington.
Univ. Calif. Publ. Geol. Sci. 35:159-276.
, J. Gray, and L. M. Huggins. 1975. Preservation of
Miocene fossils in unoxidized lake deposits, Clarkia,
Idaho. J. Paleontol. 49: 833-844.
Smiley, C. J. and W. C. Rember. 1981. Paleoecology of the
Miocene Clarkia Lake (northern Idaho) and its envi-
Middle Miocene through Pliocene History 271
rons. In: J. Gray, A. J. Boucot, and W. B. N. Berry (eds.),
Communities of the Past. Hutchinson Ross, Strouds-
burg, PA. pp. 551-590.
-. 1985. Physical setting of the Miocene Clarkia fossil
beds, northern Idaho. In: C. J. Smiley (ed.), Late Ceno-
zoic History of the Pacific Northwest (Interdisciplinary
Studies on the Clarkia Fossil Beds of Northern Idaho).
Pacific Division, American Association for the Ad-
vancement of Science, San Francisco, CA.
Smith, C. A. S., C. A. Fox, and H. Kodama. 1994. Paleosols
associated with Miocene basalts, Porcupine River,
northeastern Alaska: implications for regional paleo-
climates. Quatern. Int. 22/23: 79-90.
Smith, G. R. and W. P. Patterson. 1994. Mio-Pliocene sea-
sonality on the Snake River Plain: comparison of fau-
nal and oxygen isotopic evidence. Palaeogeogr.,
Palaeoclimatol., Palaeoecol. 107: 291-302.
Smith, H. V. 1938. Some new and interesting late Tertiary
plants from Sucker Creek, Idaho-Oregon boundary.
Bull. Torrey Bot. Club 65: 557-564.
. 1939. Additions to the fossil flora of Sucker Creek,
Oregon. Papers Mich. Acad. Sci., Arts, Lett. 24:
107-120.
Soltis, P. S. and D. E. Soltis. 1993. Ancient DNA: prospects
and limitations. N.Z. J. Bot. 31: 203-209.
and C. J. Smiley. 1992. An rbcL sequence from a
Miocene Taxodium (bald cypress). Proc. Natl. Acad.
Sci. USA 89: 449-451.
Stanton, R. J., Jr. and J. R. Dodd. 1970. Paleoecologic tech-
niques—comparison of faunal and geochemical
analyses of Pliocene paleoenvironments, Kettleman
Hills, California. J. Paleontol. 44:1092-1121.
Taggart, R. E. and A. T. Cross. 1990. Plant successions and
interruptions in Miocene volcanic deposits, Pacific
Northwest. In: M. G. Lockley and A. Rice (eds.), Vol-
canism and Fossil Biotas. Geological Society of Amer-
ica, Boulder, CO. Special Paper 244, pp. 57-68.
and L. Satchell. 1982. Effects of periodic volcanism
on Miocene vegetation distribution in eastern Oregon
and western Idaho. Proc. Third North Am. Paleontol.
Convention 2: 535-540.
Thelin, G. P. and R. J. Pike. 1991. Landforms of the conter-
minous United States—A Digital Shaded-Relief Por-
trayal. U.S. Geological Survey, Reston, VA. Miscell. In-
vest. Ser. Map 1-2206.
Thomasson, J. R. 1978a. Epidermal patterns of the lemma
in some fossil and living grasses and their phyloge-
netic significance. Science 199: 975-977.
. 1978b. Observations on the characteristics of the
lemma and palea of the Late Cenozoic grass Panicum
elegans. Am. J. Bot. 65: 34-39.
. 1979. Angiosperms from the Late Tertiary Keller
local fauna of Ellis County, Kansas. Contrib. Geol.
Univ. Wyoming 17: 59-63.
. 1980a. A fossil Equisetum sp. (family Equisetaceae,
subgenus Hippochaetae) from the Late Tertiary Ash
Hollow Formation of Nebraska. Am. J. Bot. 67:
125-127.
. 1980b. Archaeoleersia nebraskensis gen. et sp. nov.
(Gramineae-Oryzeae), a new fossil grass from the Late
Tertiary of Nebraska. Am. J. Bot. 67: 876-882.
. 1980c. Paleoeriocoma (Gramineae, Stipeae) from
the Miocene of Nebraska: taxonomic and phylogenetic
significance. Syst. Bot. 5: 233-240.
272 Late Cretaceous and Cenozoic History of North American Vegetation
—. 1982. Fossil grass anthoecia and other plant fossils
from anthropod burrows in the Miocene of western
Nebraska. J. Paleontol. 56: 1011-1017.
—. 1983. Carex graceii sp.n., Cyperocarpus eliasii
sp.n., Cyperocarpus terrestris sp.n., and Cyperocarpus
pulcherrima sp.n. (Cyperaceae) from the Miocene of
Nebraska. Am. J. Bot. 70: 435-449.
—. 1984. Miocene grass (Gramineae: Arundinoideae)
leaves showing external micromorphological and in-
ternal anatomical features. Bot. Gaz. 145: 204-209.
—. 1985. Tertiary fossil plants found in Nebraska. Natl.
Geogr. Soc. Res. Rep. 19: 553-564.
-, R. J. Zakrzewski, H. A. Lagarry, and D. E. Mergen.
1990. A Late Miocene (late Early Hemphillian) biota
from northwestern Kansas. Natl. Geogr. Res. 6: 231-244.
Thompson, R. S. 1991. Pliocene environments and cli-
mates in the western United States. Quatern. Sci. Rev.
10: 115-132.
(ed.). 1994. Pliocene terrestrial environments and
data/model comparisons. U.S. Geological Survey, Re-
ston, VA. Open-file Rep. 94-23, pp. 1-93.
-. 1996. Pliocene and Early Pleistocene environments
and climates of the western Snake River Plain, Idaho.
Mar. Micropaleontol. 27: 141-156.
Ward, L. W. and D. S. Powars. 1989. Tertiary Stratigraphy
and Paleontology, Chesapeake Bay Region, Virginia
and Maryland. 28th International Geology Congress
Field Trip Guidebook T-216. American Geophysical
Union, Washington, B.C.
and G. L. Strickland. 1985. Outline of Tertiary
stratigraphy and depositional history of the U.S. At-
lantic coastal plain. In: C. W. Poag (ed.), Geologic Evo-
lution of the United States Atlantic Margin. Van Nos-
trand Reinhold, New York. pp. 87-123.
Webb, S. D. 1983. The rise and fall of the Late Miocene un-
gulate fauna in North America. In: M. H. Nitecki (ed.),
Coevolution. University of Chicago Press, Chicago, pp.
267-306.
Webber, I. E. 1933. Woods from the Ricardo Pliocene of Last
Chance Gulch, California. Carnegie Inst. Wash. Con-
trib. Paleontol. 412: 113-134.
Wen, J. and T. E Stuessy. 1993. The phylogeny and biogeog-
raphy of Nyssa (Cornaceae). Syst. Bot. 18: 68-79.
Wheeler, E. A. and C. G. Arnette, Jr. 1994. Identification of
Neogene woods from Alaska-Yukon. Quatern. Int.
22/23: 91-102.
White, J. M. and T. A. Ager. 1994. Palynology, paleoclima-
tology and correlation of Middle Miocene beds from
Porcupine River (Locality 90-1), Alaska. Quatern. Int.
22/23:43-77.
White, J. M. et al. (+ 6 authors). 1997. An 18 million year
record of vegetation and climate change in northwest-
ern Canada and Alaska: tectonic and global climatic
correlates. Palaeogeogr., Palaeoclimatol., Palaeocol.
130: 293-306.
Willard, D. A. 1993. Pliocene Palynomorph Census Data
from the Pinecrest Beds (Tamiami Formation) in Qual-
ity Aggregates Phase 6 Pit, Southwestern Florida. U.S.
Geological Survey, Reston, VA. Open-file Rep. 93-563,
pp. 1-9.
, T. M. Cronin, S. E. Ishman, and R. J. Litwin. 1993.
Terrestrial and marine records of climatic and environ-
mental changes during the Pliocene in subtropical
Florida. Geology 21: 679-682.
Wolfe, J. A. 1966. Tertiary Plants from the Cook Inlet Re-
gion, Alaska. U.S. Geological Survey, Reston, VA. Pro-
fessional Paper 398-B, pp. 1-32.
. 1972. An interpretation of Alaskan Tertiary floras.
In: A. Graham (ed.), Floristics and Paleofloristics of
Asia and Eastern North America. Elsevier Science
Publishers, Amsterdam, pp. 201-233.
. 1978. A paleobotanical interpretation of Tertiary
climates in the northern hemisphere. Am. Sci. 66:
694-703.
. 1981. A chronologic framework for Cenozoic
megafossil floras of northwestern North America and
its relation to marine geochronology. In: J. M. Armen-
trout (ed.), Pacific Northwest Cenozoic Biostratigra-
phy. Geological Society of America, Boulder, CO. Spe-
cial Paper 184, pp 39-47.
. 1985. Distribution of major vegetational types dur-
ing the Tertiary. In: E. T. Sundquist and W. S. Broecker
(eds.), The Carbon Cycle and Atmospheric CO2: Nat-
ural Variations Archean to Present American Geo-
physical Union, Washington, D.C. Monograph 32,
pp. 357-375.
. 1990. Estimates of Pliocene precipitation and tem-
perature based on multivariate analysis of leaf phys-
iognomy. In: L. B. Gosnell and R. Z. Poore (eds.),
Pliocene Climates: Scenario for Global Warming. U.S.
Geological Survey, Reston, VA. Open-File Rep. 90-64,
pp. 39-42.
and D. M. Hopkins. 1967. Climatic changes
recorded by Tertiary land floras in northwestern North
America. In: K. Hatai (ed.), Tertiary Correlations and
Climatic Changes in the Pacific, llth Pacific Science
Congress, Tokyo, 1966 Symposium 25, Sendai, Sasaki,
pp. 67-76.
and E. B. Leopold. 1966. Tertiary stratigraphy and
paleobotany of the Cook Inlet region, Alaska. U.S. Ge-
ological Survey, Reston, VA. Professional Paper
398-A, pp. 1-29.
Wolfe, J. A. and E. B. Leopold. 1967. Neogene and Early
Quaternary vegetation of northwestern North America
and northeastern Asia. In: D. M. Hopkins (ed.), The
Bering Land Bridge. Stanford University Press, Stan-
ford, CA. pp. 193-206.
Wolfe, J. A., H. E. Schorn, C. E. Forest, and P. Molnar. 1997.
Paleobotanical evidence for high altitudes in Nevada
during the Miocene. Science 276: 1672-1675.
Wolfe, J. A. and T. Tanai. 1980. The Miocene Seldovia Point
flora from the Kenai Group, Alaska. U.S. Geological
Survey, Reston, VA. Professional Paper 1105, pp.
1-52.
. 1987. Systematics, phylogeny, and distribution of
Acer (maples) in the Cenozoic of western North Amer-
ica. J. Fac. Sci., Hokkaido Univ. 22:1-246.
Zubakov, V. A. and 1.1. Borzenkova. 1990. Global Palaeocli-
mate of the Late Cenozoic. Developments in Palaeon-
tology and Stratigraphy, 12. Elsevier Science Publish-
ers, Amsterdam.
General Readings
Fischman, J. 1993. Forever (in) amber. Science 262:
655-656.
Golenberg, E. M. 1994. Antediluvian DNA research. Nature
367: 692.

Holden, C. 1997. "No go" for Jurassic Park-style dinos. Sci-
ence 276: 361.
Lindahl, T. 1993. Recovery of antediluvian DNA. Nature
365: 700.
Poinar, G. O., Jr. 1993. Recovery of antediluvian DNA—
reply. Nature 365: 700.
Middle Miocene through Pliocene History 273
Poinar, H. R, R. J. Cano, and G. O. Poinar, Jr. 1993. DNA
from an extinct plant. Nature 363: 677.
Service, R. E 1996. Just how old is that DNA, anyway? Sci-
ence 272: 810.
Sykes, B. 1997. Lights turning red on amber. Nature 386:
764-765.
 EIGHT
Quaternary North American Vegetational History
1.6 Ma to the Present
CONTEXT SUMMARY
The Quaternary Period encompasses the Pleistocene and
the Holocene or Recent Epochs (Fig. 7.2). The date used for
the beginning of the Pleistocene depends upon which glob-
ally recognizable event is selected as representing a signif-
icant break with the preceding Pliocene Epoch. Candidates
include the Gauss-Matuyama magnetopolarity boundary
(-2.8 Ma; see Quaternary International, 1997); the initia-
tion of widespread permafrost, a frigid Arctic Ocean, and
rapid glaciation in the high northern latitudes (-2.4 Ma;
Shackleton and Opdyke, 1977; Shackleton et al., 1984); or
the African Olduvai paleomagnetic event between 1.87 and
1.67 Ma. The transition from hothouse to icehouse condi-
tions was gradual, but the Pleistocene is typified at Vrica,
Italy, as beginning at -1.67 Ma (Aguirre and Pasini, 1985;
Richmond and Fullerton, 1986; oxygen isotope stage 62),
and that is the date used here. In the conterminous United
States the Elk Creek till of Nebraska is 2.14 m.y. in age
(Hallberg, 1986), and the onset of the full ice age is repre-
sented by the onset of repeated glaciations at -850 Kya
when glaciers extended down the Mississippi River Valley
(Fig. 8.1). Subsequently, glacial-interglacial conditions
fluctuated until the latest retreat at -11 Kya that began the
Holocene or Recent Epoch.
The chronology of ice age events began with the publi-
cation of Louis Agassiz's (1840) Etudes surles Glaciers. In
the absence of evidence to the contrary, a single glacial ad-
vance was envisioned as blanketing the high latitudes. In
the 1940s Willard E Libby at the University of Chicago per-
fected the technique of radiocarbon dating, and Flint and
Rubin (1955) applied this methodology of "isotopic clocks"
to establishing the absolute chronology of drift deposits
from the eastern and midwestern United States. Their ra-
diocarbon dates showed evidence of two or more times of
continental-scale glaciations; older organic material was
274
"radiocarbon inert" and beyond the ~40-Ky range of the
technique. A standard chronology eventually became es-
tablished for North America that included four major
glacial stages (Nebraskan, oldest; Kansan; Illinoian; and
Wisconsin) separated by four interglacials (Aftonian, old-
est; Yarmouth, Sangamon, and the present Holocene).1
Then, long cores from the Atlantic Ocean documented at
least nine glacial stages (Phleger et al., 1953), and Emiliani
(1955) discovered seven complete glacial-interglacial cy-
cles in a core from the Caribbean based on 18O/16O mea-
surements. A total of 18-20 stages is presently recognized
for the past 1.6 m.y. Study of another Caribbean core
showed an eccentricity rhythm of 100 Ky (Broecker and
van Donk, 1970; Fig. 8.2), while a core from the Indian
Ocean revealed tilt and precession cycles (Hays et al.,
1976; Fig. 8.3). It is now known that between 2.5 and 0.735
Ma (Matuyama magnetic chron) glacial cycles in the north-
ern latitudes followed the tilt variation of -41 Ky; since that
time they have followed the 100-Ky eccentricity variation.
Toward the middle latitudes it is the 41-Ky cycle that has
been primary since -735 Kya. A close correspondence has
been shown between high sea levels and periods of maxi-
mum insolation in the Milankovitch cycles (Shackleton,
1987). The climatic pattern involves slow buildup to
glacial conditions (stadials), followed by abrupt switches
(terminations) to warmer conditions at 100-Kya intervals
(interstadials or interglacials; Fig. 8.4). The Greenland and
Antarctic ice cores subsequently provided evidence of
small rapid fluctuations of 7-12 Ky (Heinrich events; Fig.
8.4) and 1 Ky to -100 years (D-O events; Fig. 8.5). Climatic
instability is particularly pronounced at the onset of deglac-
iations. Correlation of terrestrial vegetational changes on the
scale of D-O events is complicated by distance from the
coast (continentality), topographic diversity, and individ-
ual lag times for migrating populations of various species
that can range from decades to centuries (Davis, 1989).

On the basis of marine oxygen isotope, Greenland ice
core, and paleomagnetic data, the Quaternary may be sub-
divided into early, middle, and late epochs. The Early Qua-
ternary extends from 1.6 Ma to the Matuyama-Brunhes
geomagnetic polarity reversal at 780 Kya; the Middle Qua-
ternary extends from 780 Kya to oxygen isotope substage
5e, corresponding to an age of 122 Kya; and the Late Qua-
ternary extends from 122 Kya to the present. The latter is
divided into the oxygen isotope stages shown in Fig. 8.5.
Other time units encountered in the literature and their ap-
proximate duration are:
Sangamon interglacial: 122 Kya (132 Kya; Richmond
and Fullerton, 1986) to 80 Kya, oxygen isotope stage
5e-a
Early Wisconsin: 80-28 Kya, stage 4
Middle Wisconsin: 28-23 Kya, stage 3
full-glacial Late Wisconsin: 23-16.5 Kya [18 Kya], stage 2
Quaternary North American Vegetational History 275
Figure 8.1. Extent of glacial and periglacial
zones of North America at present and at the
last glacial maximum. The solid line border-
ing the continent represents the approximate
position of the shoreline during the full Wis-
consin glaciation. The southern extension of
discontinuous permafrost along the Ap-
palachian high summits to the Great Smoky
Mountains in the Late Wisconsin is based on
Pewe (1983). Reprinted from Williams et al.
(1993, fig. 5, and references cited therein)
with the permission of M. A. J. Williams.
late-glacial Late Wisconsin: 16.5-12 Kya
Early Holocene: 12 Kya to 8.9-8.5 Kya, stage 1
Middle Holocene: 8.9-8.5 Kya to 4 Kya
Late Holocene: 4 Kya to the present
In North America there were two main sites of ice accu-
mulation. The Laurentide ice sheet was the largest and it
spread from a central dome located over Hudson Bay; the
Cordilleran ice sheet formed in the northern Rocky Moun-
tains of British Columbia. The first estimate of maximum
ice thickness (by Cleveland, Ohio, geologist Charles Whit-
tlesey, 1868 was ~1.8 km (~1 mi) for the Laurentide conti-
nental glacier. More recently a value of 3.5 km has been
suggested, but Peltier (1994) interprets sea-level and crustal
rebound data as indicating a thickness of ~2 km (see tech-
nical comment by Edwards, 1995; response by Peltier,
1995). The most areally extensive glaciation was the Ne-
braskan because till and outwash sediments of that age ex-
276 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 8.2. Variation in oxygen-isotope ratios from

Caribbean core V12-122 showing 100-ka eccentricity
cycle. Roman numerals I-VI indicate the six intervals of
rapid deglaciation (terminations). Reprinted from Imbrie
and Imbrie) (1979, fig. 38, and references cited therein)
with the permission of Enslow Publishers, Inc.

tend beyond the cover of later advances (Bowen, 1978).
The two centers were frequently confluent, and at the latest
glacial maximum at 18 Kya ice covered -40 million km2
(30% of the Earth's surface) or ~3 times more than the pres-
ent 15 million km2. During the last or Late Wisconsin
episode of glaciation, the Laurentide ice sheet reached its
farthest southern extension as the Des Moines lobe in cen-
tral Iowa (41° N latitude) at -14 Kya; the maximum of the
Cordilleran ice sheet was at -15 Kya. A zone of permafrost
80-200 km wide extended along the southern edge of the
Laurentide ice sheet from the western Cordilleras to the At-
lantic coast, as shown by the distribution of ice-wedge
casts and pingos (hills that were ice covered and preserve
evidence of glacial action; Pewe, 1983). Approximately 77
million km3 or 10% of the world's water was locked up in
the ice, compared to 30 million km3 or 1.7% today, and the
full-glacial position of sea level was lower than at present
by 121 ± /5 m. This increased ocean salinity by -3%, in-

Figure 8.3. Variation in oxygen-isotope ratios from Indian Ocean cores

showing temperature changes consistent with the Milankovitch varia-
tions. Reprinted from Imbrie and Imbrie (1979, fig. 42, based on Hays et
al., 1976) with the permission of Enslow Publishers, Inc.
 Figure 8.4. (Top) Oxygen isotope record of global ice volume for
the past 450 Kya. (Bottom) Percent lithic fragments in North
Atlantic deep sea sediments for the past 70 ka showing influx
of debris from icebergs during Heinrich events. Reprinted from
Clark (1995) with the permission of Peter Clark and the American
Association for the Advancement of Science.

Figure 8.5. Oxygen isotope record from the Camp Century Greenland ice
core showing climatic optimum, Younger Dryas, and D-O oscillations. The
Atitlan spike may be the eruption of the Atitlan volcano (Los Chocoyos
ash), Guatemala, 84 Kya. The spike at -75 ka may be an eruption of Mount
Pinatubo. Reprinted from Dawson (1992, fig. 2.6, based on Johnsen et al.,
1972) with the permission of Alastair Dawson and Nature.
278 Late Cretaceous and Cenozoic History of North American Vegetation
creased global land surfaces by -15%, and doubled the
size of Florida, primarily by exposing the shallower west-
ern coastal plain (Fig. 8.1).
The growth of ice sheets and the climatic cycles they
imply are determined by a complex of interacting factors
and feedbacks. The closure of the Isthmus of Panama by
-3.5 Ma strengthened the Gulf Stream and brought increas-
ing amounts of warm water and moist air into the North At-
lantic beginning in the Late Pliocene. Atmospheric CO2
concentration was decreasing, and samples of past atmo-
sphere from air bubbles trapped in the Greenland ice cores
show a relatively low -190-200 parts per million by vol-
ume (ppmv) in the last glacial compared to the present
-351 ppmv. As noted previously, the mechanism(s) that
control the input of CO2 into the atmosphere, beyond peri-
ods of major plate reorganization, are immensely complex
and not well understood. They include regulation of CO2
release from the ocean surface through the formation and
removal of overlying sea ice, thermocline and saline strati-
fication differences, residence time associated with ocean-
water circulation, and fluctuations in the abundance and
CO2 production of marine organisms. Processes in the terres-
trial realm include expansion and contraction of carbon-
storing tundra (Gallimore and Kutzbach, 1996), boreal, tem-
perate, and tropical forests, tropical grassland, and other
vegetation types; CO2 and other reactive gas storage and re-
lease from terrestrial vegetation colonizing the continental
shelf, alternately exposed or inundated with changes in sea
level; erosion rates, which also vary with sea level along
the continental shelf; precipitation of shallow-marine car-
bonate sediments; and, to a lesser extent, diminished epi-
sodes of volcanic activity. Periodically these factors re-
duced atmospheric CO2 to a level where low insolation
phases of the Milankovitch cycles were expressed as the
increasing accumulation of winter snow surviving through
intervals of diminished summer abolation.
As the Laurentide ice sheet developed, the mountain of
glacial ice divided the westerly jet stream into two compo-
nents (Broccoli and Manabe, 1987; Kutzbach and Guetter,
1986; Kutzbach and Wright, 1985; Manabe and Broccoli,
1985). One flowed north across Alaska and the Canadian
Arctic, reducing the strength of the easterlies that prevail
there at present. The other crossed North America between
-33° N and the glacial margin at -40° N. The present win-
ter flow is centered across the northern Rocky Mountains
at ~45°-47° N and shifts northward in the summer to
65°-70° N. As the air flowed across the accumulating ice it
was cooled, carried into the North Atlantic by the wester-
lies, chilled the ocean surface, and contributed to the for-
mation of sea ice. This provided a positive feedback to
colder climates, as did increased albedo from expanding
ice and decreasing soil and vegetation cover. Using either
minimum or maximum estimates for ice thickness (1.8-3.5
km), a split in the jet stream is still produced according to
simulations generated by computer models.
In these simulations, winters at glacial maxima in
Greenland were -11-15°C cooler (Cuffey et al., 1995) than
in the Early Holocene, and summers were 2°C cooler. Adja-
cent areas to the south were ~6-8°C cooler, and toward the
lower latitudes temperatures were 2°C (CLIMAP Project
Members, 1976) to 5-6°C cooler than at present (Guilder-
son et al. 1994; Thompson et al., 1995). At elevations be-
tween 3 and 5 km in the mid-latitudes temperatures were
~5°C lower at glacial maximum than at present.
Within and marginal to the ice sheets, periods of glacial
advance correspond to times of aridity and increased lev-
els of dust appear in the marine and terrestrial record (dry,
windy, and cold). Precipitation in these regions during
glacial advances was -30% of the present; also, dust in
cores from nearby dry continental interiors, glacial out-
wash, and the exposed continental shelves was 70 times
higher than during the interglacials. The semipermanent
high-pressure anticyclone system fixed over the Laurentide
dome displaced low-pressure cyclone belts to the south.
This produced pluvial or wet phases in parts of periglacial
North America (i.e., beyond the glacial boundary), espe-
cially in the southwest under the ascending arm of the
Hadley regime and in the southeastern United States north
of -33° N, which marked the position of the jet stream or
polar front (Delcourt and Delcourt, 1984). During glacial
intervals south of -33° N, the region across the southeast-
ern United States was somewhat drier than areas to the
west (Barry, 1983; Kutzbach and Guetter, 1986) and the
flow of the easterlies brought dry winds to the Pacific
Northwest.
The inland position of the Early Laurentide and
Cordilleran centers allowed ice to accumulate without ex-
tensive calving that would have diluted the northern
oceans and created a "drop dead" mode of thermohaline
ocean circulation. As the glacier margins neared the coasts,
such a mode did develop and this is one mechanism pro-
posed for the rapid reversals of climate expressed in the
Greenland ice cores.
The ice cores show the transition from the Late Pleis-
tocene to the Holocene at a depth of 1150 m, correspond-
ing to an age of -11 Kya (10,750 years B.P.; the end of the
Younger Dryas stadial). At that time glaciers worldwide
began to retreat, temperatures rose, dust declined, and CO2
and CH4 concentrations increased; Termination I had
begun. [For a series of papers on the Younger Dryas, see
Quaternary Science Reviews, (1993, 1995).] As the Mi-
lankovitch cycles entered a phase of increased summer in-
solation and enhanced seasonal contrast glaciers began to
rapidly fluctuate (Levesque et al., 1997), but generally re-
treat; melting and calving poured freshwater into the North
Atlantic Ocean and down the Mississippi River into the
Gulf of Mexico until the ice margin retreated beyond the
St. Lawrence drainage at -11 Kya. Early retreat led to accel-
erating feedbacks such as decreased erosion of the conti-
nental shelf as sea levels rose (less drawdown of CO2), de-
creased albedo from reduced ice and expanding vegetation,
increased biological activity and CO2 production in the

warming northern waters, reduced sea ice, and greater re-
lease of CO2 from the ocean surface. Some models indicate
that increased dust over high-albedo ice and snow surfaces
may have contributed to the abrupt periods of warmth in
the last glacial cycle (Overpeck et al., 1996). By the Holo-
cene, CO2 concentration in gas bubbles from the Greenland
cores had risen from 190 to 200 to 260 to 280 ppmv. Marine
isotope records from the North Atlantic reveal three
episodes of glacial melting at 14-12 Kya, 10-9 Kya, and at
the postglacial climatic optimum at 8-6 Kya (Jansen and
Veum, 1990; Mix, 1987; Fig. 8.6). During this mid-Holocene
warm period, also known as the hypsithermal (eastern
North America) or altithermal (western North America),
summer temperatures averaged 2-4°C warmer than at pres-
ent. About one-third of total ice volume was lost during the
first two episodes, and the associated rise in sea level is
documented by erosion terraces on the Barbados coral
reefs (Fairbanks, 1989). At ~12 Kya the rise there was -24 m
in -1 Kya, and at 9 Kya it rose 28 m over a slightly longer
time. The intervening pause corresponds to the Younger
Dryas stadial. Retreat of the Wisconsin glacier along its
southern margin was a rapid few hundred meters per year.
It had reached Hudson Bay by ~8 Kya and was essentially
at its present extent by ~6 Kya. Ice that had accumulated
over the previous -100 Kya melted in only 8 Kya, and
southern Greenland warmed by 7°C at the end of the
Younger Dryas stadial.
Between 1450 and 1850 A.D. there was a pause in the
warming trend called the Little Ice Age when MAT in the
middle to high latitides declined by 1-2°C. Severe winters
are verified by a variety of historical records, including
poor grain and grape harvests, livestock losses, naviga-
tional difficulties, and paintings of skaters on frozen Dutch
canals that rarely freeze today (15 times this century, the
last in January, 1997). Imbrie and Imbrie cite another inter-
esting consequence of this short-term fluctuation:
The legendary winter during which Washington's troops
were bivouacked at Valley Forge was actually regarded
by contemporary observers as "notably mild." In fact, if
Washington had camped at Valley Forge two years la-
ter, during the winter of 1779-80, the suffering of his
troops would have been much greater. Even by Little
Ice Age standards, that winter was "The most hard dif-
Quaternary North American Vegetational History 279
Figure 8.6. Air temperature over Greenland
for the past 40 Kya based on oxygen isotope
ratios in the Summit ice core. Reprinted from
Kerr (1993) with the permission of the Ameri-
can Association for the Advancement of Sci-
ence and the Greenland Ice Core Project.
ficult winter . . . that ever was known by any person liv-
ing." (1979, p. 183)
To the north, New York harbor was frozen solid. Ludlum
writes:
Though both the Hudson and the East Rivers were ac-
customed to freeze solidly from time to time in the
olden days, there was no record of the entire Upper Bay
congealing for a number of days. . . . [In late January]
people walked on the ice all the way from Staten Island
to Manhattan Island, a distance of five miles. . . . Heavy
loads and even large cannon were dragged across the
leeways to fortify the British position on Staten Island
which had been subject to cross-the-ice forays from
Washington's outposts in New Jersey. (1966, p. 115)
Regions within and along the margins of the glacial
boundary in North America developed wetter climates
during the interglacials as the jet stream and Hadley atmo-
spheric convection cell approached modern positions. The
northward shift of the low-pressure systems brought the
southwestern United States back under the descending
arm of the Hadley regime that, together with its inland po-
sition and rainshadow effects, restored warm dry climates
to the region during each interglacial. This high-pressure
system during interglacial times in the southwest also pre-
vented the inland penetration of lows from the Gulf of
Mexico and the Pacific Ocean. Climatic trends in the far
southeastern and southwestern United States were compli-
cated by the close proximity to the ocean, responsiveness
to variations in the flow of ocean currents, and up welling.
Areas are still rebounding from removal of the ice; for
example, the shores around Lake Superior are rising -38
cm (15 in.) per century. This postglacial adjustment to load
(isostasy) from the ice-age cryosphere is a delayed result of
the viscous and deformable nature of the underlying man-
tle (Peltier, 1996).
The Pleistocene land-mammal faunas of North America
group into two NALMAs. The older Irvingtonian has been
characterized as Mammuthus-pie-Bison (Savage and Rus-
sell, 1983). There are -120 genera from this age (1.6-0.5
Ma) and 12 new immigrants from Asia and eight from
South America (Webb, 1985). Steppe species from Asia in-
clude mammoths, musk-oxen, and microtid rodents (lem-
280 Late Cretaceous and Cenozoic History of North American Vegetation
mings, mice, voles, and related types); introductions from
South America include sloths (Notrotheriops, Eremoth-
eriuni), anteaters, opossum, and hystricognath rodents
(porcupines). By Rancholabrean time (-0.5 Ma) there were
-130 genera of land mammals in North America and 16
genera newly emigrated from Asia and nine from South
America. Of special importance were the Asian microtids,
goats, and bison that had significant ecological impact on
the vegetation (e.g., see exclusion experiments discussed in
Chapter 3, Faunas). Fluctuations in composition character-
ized North American mammal faunas during the Quater-
nary, paralleling that described for the flora. Although sug-
gesting a degree of randomness not always evident in the
fossil assemblages, the assessment by Graham and Mead is
similar to that expressed by several students of the Pleis-
tocene flora:
[V]ertebrate communities are not tightly bound aggre-
gates of species; rather, they are collections of species
randomly distributed along environmental gradients
(Whittaker, 1970). Consequently, environmental change
will not illicit a response from the entire community,
but instead each species will respond according to its
own tolerances. The species composition of a commu-
nity may therefore be constantly in flux, and significant
environmental fluctuations may cause major biotic reor-
ganizations. Modern communities are not direct ana-
logues for past ones, but changes in the distribution,
abundance, and clinal variation of individual species
can provide invaluable information about past environ-
ments. (1987, p. 371)
The extinction of most large land mammals in North
America is attributed to the rapid climatic changes of the
Late Pleistocene, particularly during the transition from
Late-glacial to Holocene times (Termination I), and to pres-
sures from newly arrived Asian hunters following the "ice-
free corridor" of western Canada. [For a series of papers on
the corridor, see Quaternary International (1996).]
The immigration of humans that were tracking herds of
large mammals from Asia across the Bering land bridge is a
new factor that later affected North American biotas
through agricultural disturbance as shown in pollen and
spore diagrams. Hunters and gatherers may have been in
Siberia by 30-35 Kya (Klein, 1975) and although the dates
are unsettled, they possibly reached North America by 30
Kya, corresponding to a glacial interval with lower sea
level. At this time the Bering land bridge was 1500 km wide
and unglaciated portions supported various graminoid-
shrub tundra associations described in Chapter 1. At -12
Kya the MacKenzie River Valley was mostly ice free and
there was another wave of immigration (Martin, 1973).
Human populations in Beringia and elsewhere in North
America were definitely established by -12 Kya (Josenhans
et al., 1997) and the spread of Paleolndian culture correlates
temporally with extinction of the North American megafau-
nas at 12-10 Kya.
In the older four-part chronology, glacials and inter-
glacials were considered about equal in duration (-175 Ky)
or the interglacials longer, but the 18O and ice-core evi-
dence now document that the principal interglacials lasted
only -12 Ky. There have been few periods within the past
130 Ky when ice volume was as low as it is at present.
Holocene temperatures reached their maximum between 8
and (-6) 4 Kya, and since that time they have shown an
overall decline. As noted by Davis for North America, "The
Pleistocene must be viewed as a cold, glaciated epoch, in-
terrupted periodically by catastrophic warm events—the
brief interglacials with climate similar to that of today."
(1976, p. 14). This raises the important and complex possi-
bility that the Earth entered another cold interval -4-6 Kya
and that, consistent with previous history, a new series of
glacial advances may be imminent. For example, during
the past 3 Ky there has been a southward displacement of
the ecotones between the tundra and boreal forest and be-
tween the boreal and lake states forest. The unresolved
question is the extent to which human-induced global
warming will delay or offset these solar-ocean circulation-
induced climatic trends. Efforts are under way to predict
future biome distributions under various future-climate
scenarios (VEMAP, 1995)
PLEISTOCENE AND HOLOCENE VEGETATION
Information on North American Quaternary vegetation is
not uniformly available geographically or stratigraphically.
Within the glacial boundary the advance of the Laurentide
ice sheet eroded underlying bedrock to a depth of -120 m
(Bell and Laine, 1985). With subsequent retreat of the ice
an irregular topography was created, augmented by numer-
ous kettleholes, which are steep-sided and round to flat
bottom depressions formed by remnants of stagnant ice.
Melting ice blocks in this poorly drained landscape left nu-
merous lakes and bogs. The depth of some lakes and the
rapid accumulation of sediments (typically 1 m/1000
years) reduced oxidation of organic material and allowed
preservation of plant and animal fossils, especially pollen,
spores (algal, fungal, bryophyte, fern and other vascular
nonseed plants), phytoliths, cuticles, epidermal fragments,
wood, diatoms, unicellular (e.g., Pediastrum) and colonial
(e.g., Botryococcus] algae, insects [Coleoptera (beetles),
Morgan, 1987; midges, Walker et al., 1991], molluscs, and
even small mammals. Vegetation developing around the
lake margins eventually filled some basins to form peat
bogs. The acidity of these bogs reduced bacterial and fun-
gal activity, further contributing to the preservation of the
fossils. The result is that much information on Late Quater-
nary vegetational history is available from lake and bog se-
quences within and marginal to glaciated North America.
In periglacial regions suitable sites are fewer, although in
the southeast there are karst lakes and sinkholes, bays,
buried soils, lagoon deposits, and some bogs, glades, and
fens (Jacobson et al., 1987, fig. 1). In the southwest, playa

Figure 8.7. Pollen diagram from the Cause Bog, Milan County, Texas. Note the reported presence of -11% Abies and
Picea pollen at the ~13-13.7 ft level and -20% Castanea pollen between the 11- and 13-ft levels Reprinted from Potzger
and Tharp (1954, fig. 1) with the permission of The Ecological Society of America.
lakes, packrat middens, and dendrochronology provide in-
formation on biotic and environmental history, especially
when interpreted within context information discussed in
previous chapters.
Sediments were mostly eroded away with each new
glacial advance, so less is known about the Nebraskan,
Aftonian, Kansan, Yarmouth, Illinoian, and Sangamon veg-
etation and terrestrial environments (Early and Middle
Quaternary) than for the Wisconsin and Holocene (Late
Quaternary). Also, improvements in the absolute dating
technique of 14C have extended the dating range back from
-40 Kya to as much as 75 Kya. Seasonal layers in lake sedi-
ments called varves are recognizable from the rapid accu-
mulation and larger particles deposited during the spring
than in the summer and winter. This allows temporal reso-
lution to 1 year in some Late Quaternary sequences, but an-
nually laminated varves are not widely available in older
Quaternary deposits. Layers of tephra (volcanic ejecta) oc-
casionally can be isotopically dated or tied to times of
known eruptions such as Mount Mazama and early Mount
St. Helens. Generally, however, the chronology of events is
less precise for the Early and Middle Quaternary than for
later times.
Methodologies
A considerable amount of information on terrestrial vege-
tational history for the Quaternary has come from pollen
and spore studies. The procedures for retrieving and pro-
cessing samples are described in numerous publications
(Faegri et al, 1989; Gray, 1965; Moore et al. 1991; Traverse,
1998; Chapter 4, Paleopalynology). Microfossils are identi-
fied by comparison with types in a reference collection,
and their representations are presented in the form of dia-
grams constructed by counting the microfossils present at
various levels. The sampling interval is commonly every
10 or 15 cm along the sediment sequence, and a constant
number (ranging from 200 to 1000 microfossils) are tabu-
lated for each level. Percentages for all types at a given
Quaternary North American Vegetational History 281
depth level constitute a spectrum, which gives an inven-
tory mostly of the prominent wind-pollinated plants pres-
ent at the time—the "pollen and spore rain" that accu-
mulated on the ancient bog or mud surface. From this
information the paleovegetation for that interval is charac-
terized. The spectra may be superimposed into a vertical
sequence to provide profiles of the abundance of each type
of microfossil through time. The collective spectra and
profiles constitute a diagram, and from it changes in vege-
tation over time can be traced.
The early pollen and spore diagrams were presented as
percentage histograms (see e.g., Fig. 8.7) that usually also
included sediment depth and occasionally sediment type.
More recent diagrams may give 14C ages and separate cal-
culations for total tree pollen (arboreal pollen or AP) ver-
sus herb and shrub pollen (nonarboreal pollen or NAP).
This ratio reveals changes in vegetation density of forest
relative to nonforest communities (e.g., from closed forest
to savanna or nonforested tundra vegetation). Pollen zones
are drawn to delineate major vegetation types such as tun-
dra, boreal forest, deciduous forest, or to characterize re-
gional climatic events such as the climatic optimum and
the Little Ice Age. When 14C dates are available, the age and
duration of each zone can be determined, the zones can be
correlated between sites to reconstruct regional vegetation,
and the migratory history of individual taxa can be traced
via isopols (e.g., Delcourt and Delcourt, 1987, 1993).
Isopolls are lines on a map that connect areas of equal per-
centages of a given palynomorph. These denote first ap-
pearances and similar percentages of pollen and spores at
different sites and can reveal the source, direction, and
pace of plant migrations (Delcourt et al., 1984; Jacobson et
al., 1987; Webb, 1988). This same information can be used
to detect the movement of human populations through
their impact on vegetation. Such evidence includes the si-
multaneous reduction of varied forest types, the correlated
appearance of charcoal layers, and the appearance of intro-
duced cultivars and weeds associated with forest clear-
ance. The classic studies of Iversen (1941) demonstrated
282 Late Cretaceous and Cenozoic History of North American Vegetation
the migration of early agriculturalists from the Fertile Cres-
cent into western Europe near the end of the last ice age.
Archeological zones and cultural periods used in anthro-
pology [Paleolithic or Early Stone Age (~2 Ma-10 Kya);
Mesolithic or Middle Stone Age (10-8 Kya); Neolithic or
Late Stone Age (8-3.5 [5.5] Kya); Bronze Age (3.5-1 [5.5]
Kya); Iron Age beginning at 1 Kya; and Historic] establish
the temporal relationship between vegetation, climate, and
human history (Dimbleby, 1985; Gray and Smith, 1962;
Williams et al., 1993, chapter 9). Accessory diagrams may
include fungal spores, diatoms, ethnobotanical remains,
and macrofossils (fruits, seeds, leaves, twigs, wood; Jack-
son et al., 1997). Chemical analyses provide data on lake
history (Hedges et al., 1982; Leopold et al., 1982).
Significant advances have been made in reconstructing
the composition, arrangement, and dynamics of paleocom-
munities from pollen and spore diagrams (Grimm, 1988).
The sources of error discussed in Chapter 4 have mostly
been taken into account, but Davis (1963) has emphasized
the need for continued improvement in the way vegetation
is inferred from fossil pollen and spore assemblages. A
problem with percentage counts is that fluctuations in the
abundance of any one entity affects the representation of
others, irrespective of their real history in the vegetation.
Also, the concentration of microfossils in sediments varies
with compaction and sedimentation rate, as well as with
actual changes in the vegetation (e.g., from dense forest to
open-ground tundra or with cultural burning of forests and
land clearing for cropland). Several methods have been de-
veloped that allow a more accurate reconstruction of the
composition and the abundance and arrangement of com-
ponent associations from pollen and spore diagrams. One
is to compare the modern pollen rain on moss polsters,
bogs, and other surfaces with the surrounding vegetation
that has been inventoried and mapped. This gives an indi-
cation of the kind of pollen spectrum produced by known
vegetation types and helps in identifying most similar
modern-day analogs for the ancient fossil assemblages
(Delcourt and Delcourt, 1985a).
By counting either a standard volume of material or in-
troducing a known concentration of exotic pollen, rather
than tabulating a fixed number of microfossils from each
level, a calculation can be made of pollen influx or micro-
fossils/unit of sediment (Benninghoff, 1962; Davis, 1965).
This removes the variable of sedimentation rate from esti-
mates of vegetation density. When several 14C dates are ob-
tained from a single section, a further calculation can be
made of absolute pollen frequency (or pollen accumula-
tion rates, PAR) in microfossils/unit of sediment/unit of
time (Davis, 1967; Davis et al., 1973; see Williams et al.
1993, fig. 8.9, for a comparison of percentage and pollen
influx diagrams from Rogers Lake, Connecticut). The re-
sults are then combined with data from studies of modern
pollen rain to better establish the qualitative and quantita-
tive relationships between various plant populations,
modern vegetation types, and their pollen and spore sig-
natures. These innovations have significantly improved
the accuracy of paleovegetation reconstructions from
pollen and spore assemblages, but each approach has its
own set of limitations that still provide only a general pic-
ture of ancient plant communities (Peteet, 1991; Prentice,
1988).
A knowledge of basic ecology and a familarity with
vegetation similar to that reflected by the various spectra
are also important in reconstructing paleocommunities
from pollen assemblages. Improved numerical methods
such as transfer functions are standardizing data analysis
and are yielding more accurate and reproducible results
(Birks and Gordon, 1985). Collectively, accessory diagrams,
modern pollen rain studies, pollen frequency, pollen in-
flux, increased understanding of ecological processes and
vegetation dynamics, and improved statistical analyses are
revealing the history of Quaternary vegetation and envi-
ronments in greater detail than for any other segment of ge-
ologic time (Bartlein, 1988; Birks, 1993; Gajewski, 1993;
Schoonmaker and Foster, 1991). Understanding this history
is enhanced by a knowledge of cause and effect mecha-
nisms discussed in Chapter 2 and context information dis-
cussed in Chapter 3. The literature is vast and numerous
summaries are available for the North American continent
(e.g., Bryant and Holloway, 1985; Delcourt and Delcourt,
1987, 1993; Heusser and King, 1988; Porter, 1983; Ruddi-
man and Wright, 1987; Wright, 1983; Wright et al., 1993;
see also papers in Huntley and Webb, 1988). The intent of
this chapter is to summarize broad-scale changes in re-
gional vegetation that produced the current version of
North American plant formations and associations from
Late Pliocene predecessors. Tracing this history reveals the
dynamic and complex nature of biotic-environmental in-
teractions. This is done within the constraint of a major
gap in the record between the latest Pliocene and the Wis-
consin.
Southeast
Little is known about Early and Middle Quaternary vegeta-
tion in the southeastern United States (Elsik, 1969; Groot,
1991). It may be assumed on the basis of events in the Late
Glacial and Holocene that during earlier cold intervals, up-
land (mostly coniferous) elements moved downward and
northern temperate elements expanded their range south-
ward. At present, Tsuga canadensis, Acer saccharum,
Fagus grandifolia, and Betula allegheniensis grow at mid-
dle to high elevations in the central and southern Ap-
palachians; Abies and Picea occur at high elevations in the
central Appalachians and sporadically only above -1500
m to the south. Pinus banksiana (jack pine) is presently a
northern species that grew in the south in glacial times.
With the onset of warmer interglacial climates, cool- to
cold-temperate elements either persisted in refugia at
higher altitudes and along river valleys with cold-air
drainage, evolved new ecotypes, disappeared from the re-

Lake Tulane, Highlands County, Florida
Selected Pollen Types
Figure 8.8. Pollen diagram from Lake Tulane, Highlands County, Florida. Reprinted
from Watts and Hansen (1994, fig.) with the permission of Elsevier Science-NL.
gion, or became extinct at the species level. At Wilcox Bluff
in Louisiana, a fluvial terrace assemblage probably dates
from the Sangamon Interglacial or Early Wisconsin (Del-
court and Delcourt, 1977, in press). The vegetation is
warm-temperate and includes Juniperus, Pinus, Alnus,
Carya, Cephalanthus (buttonbush), Fergus, Fraxinus, Ilex,
Itea (Virginia willow), Leitneria (corkwood), Liquidambar,
Liriodendron, Nyssa, Ostrya-Carpinus, Platanus, Quercus,
Salix, Tilia, and Vitis. Warm-temperate to subtropical ele-
ments (e.g., palms, Bursera, Ficus, Rhizophora) were likely
introduced elsewhere in the southeast and/or expanded
their range northward and into higher elevations during
the interglacials and migrated southward, downward, or
disappeared from the region in glacial times. The most re-
cent introduction of tropical elements into peninsular
Florida probably occurred 5-6 Kya. The north-south align-
ment of the Appalachian Mountains, climatic changes, and
continuous land connections to boreal zones to the north
and tropical zones to the south made the region both a
pathway and a refugium for plants and animals throughout
Cenozoic time. The result is the most species-rich biotic
province in North America north of Mexico.
The Late Quaternary history of vegetation in the Florida
Peninsula is preserved in numerous lakes formed in lime-
Quaternary North American Vegetational History 283
stone sinkholes. One of these is Lake Tulane in Avon Park,
south-central Florida. An 18.5-m core extends from the
present back to -50 Kya (Watts and Hansen, 1994; Fig. 8.8).
The Wisconsin portion records a series of peaks in Pinus
pollen at -14-16, 21-23, 26-28, 30-33, 36-37, and 48-51
Kya. All but the 30-33 Kya interval correlate with Heinrich
events H1-H5 (Bond et al., 1992; Grimm et al., 1993; Fig.
8.9). Alternating with pine are zones of abundant Quercus
pollen and associated herbs (Gramineae, Chenopodiaceae,
Iva, Ambrosia types). Pollen resembling Taxodium is pres-
ent throughout the profile and it becomes more abundant
during high stands of sea level. The probable scrub nature
of the pine association during cold intervals was due in
part to sea levels (and water tables) that were lower by -20
m. At glacial maximum the MAT is estimated to have been
~2°C cooler (Brunner, 1982); but if the oxygen-strontium
readings from the Barbados corals are correct (5°C lower;
Guilderson et al., 1994), it may have been cooler.
In the Late Holocene after -6 Kya, pine woods (Pinus
clausa, sand pine; P. elliottii, slash pine), oak woods (Quer-
cus laevis, turkey oak), and a flood-plain vegetation of bald
cypress (Taxodium distichum) were reestablished, but with
a herbaceous understory vegetation typical of the modern
associations. According to the Lake Tulane profile, the re-
284 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 8.9. Lake Tulane pine pollen peaks and Heinrich

events 1-5. Reprinted from Watts and Hansen (1994 fig. 5) with the
permission of Elsevier Science-NL.

cent history of interior peninsular Florida vegetation has
been pine-oak-grassland (warm intervals) alternating
with open pine woods (cold intervals) within a mosaic of
sand-dune scrub and with the present interglacial version
dating back ~6 Ky.
In northwest Florida, Camel Lake shows a similar alter-
nation between Pinus and Quercus, but there is better rep-
resentation of Carya and Fagus pollen than at Lake Tulane
to the south and at one brief interval between 12 and 14
Kya Picea is present. This is the only locality in Florida
where prehistoric spruce pollen occurs in even modest
quantity (typically 2%, one spectrum up to 8%). While
Pinus banksiana (jack pine) decreased in the southern Ap-
palachian Mountains beginning -15 Kya and was replaced
mostly by Quercus, Picea remained abundant until -12
Kya to form a spruce-oak forest as in the lower midwestern
United States. Both Fagus and Picea grow north of Florida
in the Appalachian Mountains, which suggests tempera-
tures in northern Florida were lower than at present (possi-
bly to a -5°C January mean; Watts and Hansen, 1994).
To the east, Sheelar Lake is closer to the southeastern
terminus of the Appalachian Mountains, and the fossil as-
semblage includes more mesic forest elements. At 12-14
Kya a deciduous forest was present (Carya, Celtis, Fagus,
Fraxinus, Ostrya-Carpinus, Platanus, Ulmus). After -12.7
Kya a warmer, drier climate developed and grassy openings
appeared as shown by increasing Gramineae, chenoam,
and Compositae-Asteraceae pollen. This was followed by
an increase in Quercus then Pinus pollen at -8 Kya to form
the most recent version of the modern pine (Pinus palus-
tris, long-leaf pine)-oak (Quercus laevis, turkey oak) vege-
tation. The Sheelar Lake region served as one refugium for
mesic deciduous elements during the cold intervals of the
Quaternary.
In the Tunica Hills region of Louisiana-Mississippi,
sediments dated between 25,250 and 12,500 years B.P. (Del-
court and Delcourt, in press) contain megafossils and
abundant pollen of Picea (40-70%). Abies and Larix are
absent, but they are found to the north in the central Mis-
sissippi River Valley (Royall et al., 1991). Deciduous hard-
woods are rare, and the climate during the Late Wisconsin
was cooler than at present but not "boreal" (Jackson and
Givens, 1994).
In the western part of southeastern United States an
early study was made on two bogs in central Texas (Potzger
and Tharp, 1943, 1947, 1954). The most significant result
was the report of Abies and Picea pollen totaling 11% in
the lowermost level of the Cause bog (Fig. 8.7). The present
vegetation is an oak (Quercus stellata, post oak)-hickory
(Carya texana) association, or post-oak savanna, with abun-
dant grasses in the understory. funiperus and Prosopis are
also abundant, but they have increased in modern times
from overgrazing. The presence of two northern conifers
contributing 11% pollen implies a substantial boreal forest
in the region. The nearest populations of Abies and Picea
today are several hundred miles to the northeast in the

Appalachian Mountains and to the west in the southern
Rocky Mountains. Brown (1938) reported cones, twigs, and
wood of the boreal gymnosperms Larix, Picea, and Thuja
from Pleistocene deposits in Louisiana; Wilson (in Davis,
1946) found pollen of Picea in peats from northern Florida.
Collectively, these studies suggested that during glacial pe-
riods, climates had changed so dramatically in the south
that a boreal forest extended from the glacial margin south
to the Gulf Coast from Texas to Florida (Deevey, 1949).
In the absence of further information from the fossil
record of the Gulf Coast, it was difficult to evaluate this
model. The first direct evidence that revisions were neces-
sary came from a restudy of the Texas material (Graham
and Heimsch, I960).2 Abies pollen was not encountered,
Picea pollen ranged from absent to 1.5% to a maximum of
3% (6 grains/200 count at one level), and pollen of Cas-
tanea was not recovered. The closest stands of Castanea
(C. pumila, Allegheny chinquapin) are in northeast Texas
where numerous eastern deciduous trees, shrubs, and
vines have their western limits before encountering the
warm-temperate to arid habitats and deep sand to rocky
soils of the south and west. This is an important floristic
boundary for many elements that shifted their range with
climatic changes of the Quaternary. Eastern deciduous for-
est representatives here include Alnus serrulata, Aralia
spinosa (devil's walking stick), Asimina triloba (pawpaw),
Betula nigra, Carpinus caroliniana (American hornbeam),
Cercis canadensis (redbud), Chionanthus virginicus (fringe-
tree), Diospyros virginiana (persimmon), Euonymus atrop-
urpureus (eastern wahoo), Fagus grandifolia, Gymnocladus
dioicus (Kentucky coffee tree), Juglans nigra, Liquidambar
styraciflua, Madura pomifera (osage orange), Magnolia
grandiflora, M. virginiana, Myrica cerifera (wax myrtle),
Ostrya virginiana (eastern hop hornbeam), Oxydendrum
arboreum (sourwood), Persea borbonia (red bay), P. palus-
tris (swamp bay), Platanus occidentalis, Populus deltoides
(eastern cottonwood), Pyrus angustifolia (crabapple),
Rhamnus caroliniana (Carolina buckthorn), Sabal minor
(bush palmetto), Sambucus canadensis (American elder),
Sassafras albidum, Stewartia malacodendron (Virginia
stewartia), Styrax grandifolius (snowbell), Toxicodendron
vernix (poison sumac), Vaccinium arboreum, and species
of Acer, Aesculus, Bumelia, Carya, Celtis, Cornus, Cratae-
gus, Fraxinus, Gleditsia, Halesia, Ilex, Morus, Nyssa,
Prunus, Rhus, Salix, Tilia, Ulmus, Viburnum, and Zan-
thoxylum.
Megafossils identified as Thuja occidentalis (northern
white cedar) in the Louisiana deposits were later found to
be Chamaecyparis thyoides (southern white cedar), while
Picea and Larix were confirmed in sediments dated at
12,740 14C years (Delcourt and Delcourt, 1977, in press).
These data indicate significant but less profound change in
the vegetation of the region than first described. Boreal el-
ements migrated southward partly along the Mississippi
River Valley system with cold-air drainage and increased
moisture from fog. This brought stands of Picea within
Quaternary North American Vegetational History 285
pollen dispersal range of central Texas but not extensive
boreal forests along the entire Gulf Coast.
Letters from John Potzger to B. C. Tharp provide some
interesting insight into the Texas studies. In a letter dated
May 5, 1942, Potzger notes, "My one fond hope is that the
lowest levels will show pollen of northern species, if pos-
sible spruce and fir," and upon receiving word that the sam-
ples had been sent, he writes (November 19, 1942), "I can
hardly wait for the arrival of the samples. I hope we find
Picea in the bottom layers, that would be one of the most
outstanding finds of the decade." Only spruce was initially
identified, but in another letter (May 12, 1943) he "Had
word from my friend today with respect to the pollen from
Patschke bog which I had interpreted as Picea. He agrees
with me on some but thinks that Abies is also represented."
After that Potzger also began recognizing Abies in the ma-
terial. In these early days, the morphologic characteristics
that distinguish the winged pollen of Abies, Picea, and
Pinus were not well known and, indeed, confusion about
them led to misidentifications in other material (Potzger,
1944).3
Recent studies have provided additional information
on the late-glacial and postglacial history in the southwest-
ern part of the southeastern United States (Bryant, 1977;
Bryant and Holloway, 1985; Bryant and Shafer, 1977; Hol-
loway et al, 1987; Larson et al., 1972; Fig. 8.10). Peats
dated as older than 15,590 ± 250 years B.P. (late full glacial)
confirm the presence of Picea within pollen range of the
Texas sites. Deciduous elements from northeast Texas ex-
tended to the southwest along rivers, lakes, and in swampy
areas (Acer, Alnus, Betula, Carya, Cornus, Corylus, Fraxi-
nus, Populus, Tilia]. Drier areas supported oak savanna
(Quercus and Gramineae) and grassland during the full
glacial in the southern High Plains (Hall and Valastro,
1995). The mean July temperature at the end of full-glacial
times is estimated at ~2-3°C cooler than the present 27°C,
and rainfall as about the same (Dillon, 1956) or slightly
greater than the present 810 mm (Bryant, 1977). Vertebrate
records that extend back ~20 Ky on the adjacent Edwards
Plateau to the west indicate mean summer temperatures
~6°C lower than the present 27°C at glacial maximum and
within 2-3°C of present values at -13 Kya (Toomey et al.,
1993). Noble gases (Ar, Kr, Xe) dissolved in groundwater
from the Carizzo Aquifer in south Texas indicate a MAT
~5°C lower than at present for the last glacial maximum
(Stute et al., 1992). Precipitation at full glacial was less
than the present 810-510 mm (east to west gradient). In
the late glacial many of the deciduous elements disap-
peared, and oak savanna with juniper (funiperus ashei) and
mesquite (Prosopis glandulosa) was established.
The history of this ecotonal region involved introduc-
tion of Picea and mesic deciduous vegetation from the
north during cool moist intervals, the establishment of oak
savanna or an oak-hickory (Carya illinoiensis) association
in warmer drier times, and possibly the incursion of arid
elements from the west during periods of maximum tern-
286 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 8.10. Pollen diagram from the Boriack Bog, Lee County, Texas. Reprinted from Bryant (1977, fig. 2) with the per-
mission of the American Association of Stratigraphic Palynologists Foundation.

perature and dryness. The area is presently bordered to the
west and southwest by Juniperus and Prosopis woodland,
shortgrass prairie, and the Chihuahuan Desert. The present
high-pressure system over the southwest mostly prevents
moist air from the Gulf of Mexico from penetrating into
central Texas and westward, but during glacial times the
presence of the ascending low-pressure arm of the Hadley
convection cell allowed moisture from the Gulf of Mexico
to reach further inland.
In the central Appalachians more sites are available
and the vegetational history is known in greater detail. The
area is further north, closer to the Wisconsin glacial bound-
ary, and many of the communities are arranged along rela-
tively steep altitudinal gradients. As a result, changes in
climate and other factors produce complex alterations in
vegetation that are reflected in the pollen profiles. At the
last full-glacial maximum, a form of alpine tundra was pres-
ent at the highest elevations in the central Appalachian
Mountains and to the north (Chester County, Pennsylvania,
Martin, 1958; Buckle's Bog, Allegheny Plateau of Mary-
land, Maxwell and Davis, 1972; Potts Mountain Pond, Vir-
ginia, Watts, 1979). This is inferred from low percentages
of tree pollen, abundant Cyperaceae and other herbaceous
plants, and a high influx of minerals suggesting open
ground (Delcourt and Delcourt, 1986; Mills and Delcourt,
1991). The present occurrence of exposed unglaciated
granitic outcrops and invasion of sparsely vegetated sur-
faces by pioneer species of Cyperaceae, Gramineae, and
other herbs and shrubs with pollen indistinguishable from
tundra species (e.g., tree vs. dwarf species of Betula; Ed-
wards et al., 1991) illustrates the difficulty in characteriz-
ing the ancient community. The characterization of this
paleovegetation as equivalent to high Arctic tundra is spec-
ulative and has unlikely implications (e.g., for light re-
gimes; Chapter 1, Tundra). The composition and prevail-
ing physical conditions are more suggestive of an alpine
version of low Arctic tundra or tundra-spruce parkland.
At lower elevations there was a version of the Appalachian
coniferous forest (Abies, Picea), including additional ele-
ments of the boreal forest growing south of their present
distribution (e.g., Pinus banksiana, jack pine; Delcourt and
Delcourt, 1985b; Watts, 1970; Whitehead, 1981; see also
distribution in Elias, 1980). Fossil pollen of P. banksiana
probably can be recognized to species level based on size
and morphology (e.g., Ammann, 1977), but megafossils at
several sites further confirm the presence of P. banksiana
in periglacial southeastern and midwestern United States
during the latest full glacial. The record implies that the

Appalachian coniferous forest expanded and contracted,
and that boreal forest elements appeared and disappeared
repeatedly from the region during the glacial-interglacial
cycles of the Quaternary. The disjunct distribution of other
northern species in the central and southern Appalachian
highlands, which cannot be identified to species level from
pollen (e.g., Betula populifolia; Elias, 1980), could be the
result of long-distance dispersal any time during the Late
Cenozoic; but target areas were certainly increased during
the numerous cold intervals.
The dynamic nature of the vegetation in the southeast-
ern United States is documented by studies at Anderson
Pond (elevation 300 m, Middle Tennessee; Delcourt, 1979;
Delcourt and Delcourt, 1985b). This site presently supports
mixed mesophytic forest, but at 19 Kya it was Appalachian
coniferous forest. If interglacials are atypical "catastrophic
warm events" occurring regularly every 100 Ky during the
predominantly glacial climates of the past 1.6 Ma (Davis,
1976), then the present extensive deciduous forest is excep-
tional in eastern North America, which was covered most
of the time by versions of an Appalachian coniferous -
boreal forest.
A pollen assemblage from the Piedmont region of
northeastern Georgia, dated at -26-22 Kya (just before the
Wisconsin glacial maximum), was rich in Pinus and Quer-
cus and had lesser representation of Abies, Picea, and
Carya (Jackson and Whitehead, 1993). At glacial maximum
the Carolina Bays on the coastal plain were surrounded by
spruce and northern pine and had few representatives of
the deciduous forest (Whitehead, 1981). During the inter-
glacials southern yellow pine was dominant. Thus, the At-
lantic Coastal Plain was not a refugmm for the deciduous
forest in glacial times.
During these cold periods the mixed mesophytic and
other associations of the deciduous forest formation did
persist in at least three refugia identified from the pollen
and associated megafossil record: Nonconnah Creek in the
blufflands of southwestern Tennessee (Delcourt et al.,
1980); Goshen Springs in south-central Alabama (Delcourt,
1980); and, as previously noted, Sheelar Lake in northern
Florida (Watts and Hansen, 1994; Watts and Stuiver, 1980).
The deciduous forest spread from these Gulf Coast refugia
during warm intervals in patterns that seem clear for the
Holocene (Davis 1981b, 1983; Delcourt and Delcourt,
1975). However, the few number of refugia identified to
date make provisional any detailed reconstruction of the
direction and chronology of migrations during full glacial
and earlier times (Ritchie, 1987).
In the late-glacial (16.5-12.5 Kya) pollen of the Pinus
banksiana type began to disappear, the southern limit of bo-
real elements moved northward, and there was an increase
in pollen of mesic deciduous species. TAMS 14C dates from
Virginia indicate warming in the south may have begun by
17 Kya (Kneller and Peteet, 1993). During the climatic opti-
mum or hypsithermal interval (8500-4 Kya), pollen of drier
habitat species of Quercus and Carya increased and these
Quaternary North American Vegetational History 287
trees were probably present on the coastal plain, but by 5
Kya they were replaced there by the modern pine woods as-
sociation. Such interchanges must have occurred fre-
quently during the Quaternary, but the latest version of the
southern yellow pine woods association along the coastal
plain dates from ~5 Kya. Castanea dentata extended its
range into the central Appalachian Mountains to form an
oak-chestnut forest, and Pinus echinata (short-leaf pine)
moved from the south and west into the Ozark region of Ok-
lahoma and Missouri (Delcourt and Delcourt, 1991).
Evidence for early human occupation in eastern North
America comes from Meadowcroft Rockshelter in Pennsyl-
vania. Ages ranging from 12 to 20 Kya have been suggested,
but estimates from well-dated sites favor occupation after
-12 Kya; there was little impact on regional vegetation
until the last few hundred years (McAndrews, 1988; Fig.
8.11). The influence of Amerindian populations on the
vegetation begins in the Late Holocene with the recovery of
pollen of Zea mays (2 Kya; Whitehead, 1965; Whitehead
and Oaks, 1979; Whitehead and Sheehan, 1985) and re-
mains of Lagenaria siceraria (gourd), Cucurbita pepo
(squash), and Phaseolus vulgaris (bean; Chapman et al.,
1982; Cridlebaugh, 1984). Also present are pollen of plants
indicating open habitats associated with cultivation (Am-
brosia, Chenopodium type, Iva, Plantago, Rumex] and in-
creased charcoal resulting from land clearing and food
preparation (Delcourt, 1987; Delcourt et al., 1986).
Northeast
Glaciers extended south to -40° N at 18 Kya, then melting
began and intensified between 12 and 10 Kya. The general
sequence of vegetation change near the glacial boundary
was outlined early from bog and lake sequences in Con-
necticut (Deevey, 1939; see also Leopold, 1956). The pres-
ence of tundra along the margin of the Laurentide ice sheet
was considered likely, but percentage tabulations were not
adequate to detect treeless vegetation receiving pollen
input from adjacent forests. Either tundra or park tundra
(herbs with pine and spruce pollen, Peteet et al., 1993;
spruce needles, Watts, 1979) has been documented during
the full glacial in Maryland (Maxwell and Davis, 1972),
Pennsylvania (Martin, 1958), and New England (Gaudreau
and Webb, 1985); the presence of an 80-200 km wide zone
of permafrost fringing the ice suggests this vegetation was
typical of the ice margins (Miller, 1992). A bryophyte flora
from Connecticut dated at 13.5 Kya reveals tundra similar
to that now found along the northern margin of the decid-
uous forest (Miller, 1993). By 11.5 Kya the region was bo-
real woodland. In several places megafossils of tree species
were recovered, and whether this vegetation should be
characterized as tundra typical of Arctic regions is open to
question (Birks, 1976).
After the tundra-park tundra phase, Deevey (1939) rec-
ognized three zones (A, B, C) with several subzones (e.g., A-
1, A-2) to describe the sequence of vegetation and implied
288 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 8.11. Pollen diagram from Crawford Lake, Ontario, Canada. Reprinted from McAndrews (1988,
fig. 5) with the permission of J. H. McAndrews, Kluwer Academic Publishers, and the Geological Survey
of Canada.
climates for the region. Although details vary at different
places in the northeast and adjacent regions (e.g., Adiron-
dacks, Jackson and Whitehead, 1991; New Jersey, Peteet et
al, 1990, 1993; Pennsylvania, Watts, 1979, Fig. 8.12; and
Whitehead et al., 1989), the sequences are similar and begin
with a spruce-fir forest with birch locally abundant, invad-
ing the tundra as climates warmed and the glacial margin re-
treated (zone A-l). Spruce and fir reached a maximum (A-2)
then declined. As climate warmed to the climatic optimum,
pine increased (B); then with amelioration the deciduous
forest became established through oak and hemlock (C-l),
hickory (C-2; oak and hemlock decline), and chestnut-hem-
lock-oak phases (C-3). Evidence of human disturbance ap-
pears late in C. These phases were later modified (Davis,
1969; Peteet et al., 1993) and are summarized in Table 8.1.
At the western limits of the deciduous forest in eastern
and central Ohio there is a similar sequence from spruce,
to pine, and then deciduous trees (Shane, 1975, 1987,
1989a; Shane and Anderson, 1993). Slightly further to the
west on the till plains of Ohio and Indiana the history be-
gins with an open spruce forest tundra between 15.5 and
13.5 Kya. There follows a deciduous open woodland (13.5-
11 Kya), reversal to a cooler period (increase in spruce) cor-
related with the Younger Dryas event (11-10.3 Kya), and
rapid warming between -10.3 and 4 Kya (decline of hem-
lock, jack pine, red pine, white pine). After 10 Kya a mixed
deciduous forest was established; continued warming led
to the development of an open oak forest between 8 and 4
Kya (King, 1981; Shane, 1987; Webb et al., 1983). Transfer
function equations give an estimate of January mean tem-
peratures rising from -14 to -2°C and July mean tempera-
tures from 19.4 to 23.5°C between 13 Kya and 4500 years
B.P. In the latter part of this period, the June mean temper-
ature cooled to the present 22.5°C. Annual precipitation
varied from -660 to -750 mm (Shane, 1989b).
It is tempting to attribute all these regional and local
changes in vegetation to climate, but the history is more
complex. At -4800 years B.P. hemlock declined rapidly,
then recovered beginning -3400 years B.P. Other associates
of hemlock with comparable ecological requirements did
not show a similar pattern, and it is probable that the de-
cline was due to epidemic disease (Allison et al., 1986;
Davis, 1981a). Considering the recent history of species of
Castanea (chestnut) and Ulmus (elm), it is not surprising
that pollen diagrams record similar events in the past (Pat-
terson and Backman, 1988). Also, the individual migration
rates of organisms cause them to enter habitats at various
times after displacement and to form different alliances
with ecologically compatible species through time (Davis,
1981b, 1983; Webb, 1987). Past changes in the distribution
of plant formations and associations are most frequently
the consequence of climate; but if individual elements de-
 Quaternary North American Vegetational History 289

Figure 8.12. Pollen diagram from Crider's Pond, Pennsylvania. Reprinted from Ritchie (1987, fig. 4.5)
with the permission of Cambridge University Press.

viate from the pattern, several explanations are available.
These include disease, differential migration rates, indi-
vidual accommodation to stages of soil formation, and se-
lective anthropogenic impact.
Further to the north, arid permafrost conditions existed
in Greenland, Spitsbergen, and the Arctic Islands through-
out the Quaternary, and persist today as the High Arctic
polar desert. Pollen from Sangamon deposits gives an in-
sight into the vegetation and climate at the end of the last
interglacial into the Early Wisconsin. At East Milford
Quarry in Nova Scotia (>50 Kya) the sequence is hardwood
forest (Acer, Carya, Fagus, Quercus, Tilia, Ulmus; climates
at least as warm as at present), a mixed forest (Abies, Picea,
Betula; cooling), and coniferous forest (Abies, Picea; cold;
Mottetal.,1982).
Very little of present-day Canada was ice free at the last
glacial maximum, so records begin progressively later to-
ward the glacial center and preserve events primarily from
the late glacial and Holocene (Ritchie, 1987). The Lauren-
tide ice sheet separated from the northern Appalachian glac-
iers at -11 Kya and northern Maine was subsequently ice
free. Beginning -14 Kya melting in the Gulf of St. Lawrence
region was calving ice into the St. Lawrence Estuary, and by
-13-11 Kya vegetation was recolonizing the coastal areas of
the Gaspe Peninsula and Anticosti Island (Richard, 1994a).
Blocks of stagnant ice remained in central Quebec until -6
Kya. As the conifer-dominated Appalachian forest spread
northward with the retreating ice sheet and as mesic decid-
uous elements expanded from refugia, a variably defined
lake states northern deciduous-boreal coniferous forest be-
came widespread, especially in extensive areas of low to
moderate physiographic relief. This transitional vegetation
shifted northward during the climatic optima in all of the
various interglacials, most recently at -7 Kya.

Table 8.1. General sequence of pollen zones and inferred vegetation and climate for New England region.
Pollen Zones Years B.P. Vegetation Climate
C-3 Present Oak— Chestnut - Hemlock Moister, Cooler
(Ambrosia, Rumex, other weeds)
C-2 Oak, with hickory Warm -dry
C-l 7,900 Oak-hemlock (Ambrosia) Warm -moist
B 7,900-8,100 Pine maximum, with oak Warm -dry
A-4 9,100-10,200 Spruce-fir with larch, birch, alder Cooler-moister (Younger Dryas)
A-3 10,200 Spruce -oak, with pine Warmer— drier
A-2 11,700 Spruce, with oak
A-l 11,700-12,150 Spruce-fir-birch Cool, moist
"T" 12,150-14,300 Tundra Cold, dry
Adapted from Davis (1969), Deevey (1939), Leopold (1956), and Peteet et al. (1993). T = tundra.
290 Late Cretaceous and Cenozoic History of North American Vegetation
Early studies on the late-glacial and postglacial vegeta-
tion of eastern Canada were made by Terasmae (e.g., 1958,
1960, 1967) and more recently by Anderson (1985), Gajew-
ski et al. (1993), Payette (1993), Richard (1993, 1994a,b),
Ritchie (1987), and others. The sequences show an early
nonarboreal stage prior to -10 Kya of Cyperaceae, Gram-
ineae, dwarf birch (Betula glandulosa), and Salix, which
was interpreted as a desert tundra that changed to an herb-
rich tundra and then to a shrubby (dwarf birch) tundra.
There followed an afforestation stage with increasing AP
(Abies, Picea, Pinus banksiana, Betula, Populus} between
-11 and 8.5 Kya (beginning later in the highlands at ~8 Kya
and lasting until 5 Kya). There were brief reversals in the
trend toward greater forest vegetation as shown by pollen
and midge evidence from the Canadian Maritime provinces
(Levesque et al., 1993). There a cold interval occurred be-
tween 11,160 and 10,910 years B.P., just before the Younger
Dryas event. After that the canopy gradually closed with the
addition or increase in Larix, Picea, Fraxinus, Ostrya, Quer-
cus, and Ulmus; the modern boreal forest was established
across broad regions of central Canada. It extended farthest
north most recently at -6 Kya (Webb, 1987).
The chronology of these events in eastern to central
Canada is generally time-transgressive from south to north
and from east (coastal) to west (interior). The present cline
in vegetation from coastal (and lake margin) cold-temper-
ate deciduous forest to boreal forest westward to tundra
northward in east central Canada was established by ~3
Kya (Richard, 1994b). High-resolution pollen studies an-
chored by 14C dates further demonstrate that the invasion
of these vegetation types was rapid. The change from tun-
dra to closed canopy Picea mariana (black spruce) forest in
central Canada occurred in 150 years between 5 and 4 Kya
at rates estimated at 200 km/century (MacDonald et al.,
1993). In the past 3 Ky, tundra in northern Quebec has in-
creased at the expense of forest (Gajewski et al., 1993). In
the far north around Ungava Bay in northern Quebec, tun-
dra persisted throughout the Holocene.
The spread of tree species from the south followed two
principal routes along the sides of the Laurentide ice sheet:
the Labradorean pathway (boreal deciduous trees favored
by humid environments) and the Hudsonian pathway
(Picea mariana, black spruce). Many of these elements il-
lustrate again one of the important results of recent re-
search on Quaternary vegetational history: the associations
are temporal in nature. Pinus banksiana is in equilibrium
with the present climate and is not expanding its range,
Picea glauca is in equilibrium in eastern Quebec but it is
spreading along Hudson Bay, and Abies balsamea is ex-
panding in the northern James Bay region (Payette, 1993).
Plains and Upper Midwest
To the west in the Plains area there are few plant assem-
blages of Early to Middle Quaternary age (Baker and Wain,
1985; Watts, 1983). In southwestern Kansas and adjacent
Oklahoma pollen from possibly Illinoian full-glacial de-
posits are rich in Pinus and have smaller amounts of Picea,
Gramineae, Compositae, Artemisia, and rarer Alnus, Be-
tula, Carya, Fraxinus, fuglans, Quercus, and Salix. The
vegetation is interpreted as a pine savanna with spruce in
local moist habitats and deciduous trees confined mostly
to streamsides (Kapp, 1970). Toward the warmer Sanga-
mon interglacial Pinus decreases, Picea disappears, and
grasses and other herbs increase (drier and/or warmer).
Near the beginning of the Wisconsin, Pinus and Picea re-
turn (moister and/or cooler).
In the Wisconsin stage of northeastern Kansas Picea
was prominent between 24 and 14 Kya, followed by an in-
crease in mesic deciduous elements (Carpinus, Fraxinus,
Quercus, Ulmus} and establishment of the modern grass-
land by ~10 Kya (Griiger, 1973). Megafossils from the Mid-
dle Wisconsin of Missouri confirm at least some of the pine
as Pinus banksiana (King, 1973). By the Late Wisconsin
and near the glacial maximum (-20 Kya) the Ozark high-
lands in west-central Missouri were covered by spruce for-
est until -14 Kya, followed by mesic deciduous elements
(Fraxinus, Ostrya, Quercus, Ulmus}. In adjacent northeast-
ern Kansas Picea had disappeared by -11.5 Kya, and at -5
Kya the modern open woodland of Quercus and Carya
with understory grasses and other herbs was established.
After -5 Kya in southern Oklahoma the vegetation fluctu-
ated between Quercus savanna, the present-day Quercus-
Carya woodland (extending into central Texas), and pine
woodland probably extending inland from the east Texas-
Gulf Coast region (P. echinata, short leaf; P. taeda, loblolly;
P. palustris, long leaf). The latest Quercus-Carya wood-
land has existed for the past -2 Ky. The broad history of
vegetation in this region has been repeated shifts between
grassland and savanna-woodland (Pinus, Artemisia, Carya,
Quercus) with incursions of boreal conifers during the
coldest periods.
The northern grasslands are bordered by boreal forest to
the north and deciduous forest to the east and, as expected,
the ecotones shifted with the climatic changes of the Qua-
ternary. Among the first studies demonstrating the dynamic
aspect of these Pleistocene biotas were those of Wright et
al. (1963) and McAndrews (1966) that showed shifts in the
grassland-forest boundary in the upper Midwest. With
these shifts temporary associations are inferred that have
no modern counterparts. Spruce woodland, spruce-oak
woodland, and black ash tundra are examples (Chapter 3;
Fig. 8.13). In eastern Iowa and Missouri the vegetation be-
tween 34 and 22 Kya was an open Picea-Pinus forest simi-
lar to tundra-tree-line vegetation (Baker et al., 1986).
Pollen of Pinus then declined, Picea remained about the
same or increased, and the pollen of herbs increased, prob-
ably reflecting one of the many stadials of the Wisconsin
ice sheet. One of the differences between ice margin vege-
tation in the upper Midwest and in eastern North America
at glacial maxima was that the former was tundra-park-
land (with spruce), while the latter had fewer trees. A com-
 Quaternary North American Vegetational History 291

Figure 8.1 3. Pollen diagram from Kirchner Marsh, Dakota County, Minnesota. Reprinted
from Grimm (1988, fig. 3, based on Wright et al., 1963) with the permission of Eric
Grimm, the Geological Society of America, the Geological Survey of Canada, and Kluwer
Academic Publishers.

bination oiPicea and Larix, with some Fraxinus, was pre-
sent between 14 and 12.5 Kya. This was followed by incur-
sions of Acer negundo, Corylus cornuta, and Quercus
rubra into the Picea-Larix forests; disappearance of
Picea-Larix and an increase of deciduous trees between
-11 and 9 Kya; and decline of deciduous trees with an in-
crease of grassland elements at 9 Kya. By 7 Kya grasslands
were established in western Iowa and deciduous forest
grew to the east. At the height of postglacial warming (mid-
dle hypsithermal interval, ~6 Kya), the grassland reached
its greatest eastern extent and grew beyond the present
prairie peninsula of Illinois. At Wolf Creek in Minnesota
the sequence is tundra (Antennaria, Dryas, Silene, Vac-
cinium; 20.5-14.7 Kya), a transitional shrub tundra (Alnus,
Betula, Empetrum, Salix, Shepherdia; 14.6-13.6 Kya);
Picea woodland (13.6-10 Kya); and shifts between the lim-
its of the mixed conifer (Pinus banksiana)-deciduous for-
est of the present (Acer, Carya, Castanea, Celtis, Fagus, Os-
trya-Carpinus, Tilia, Ulmus) and grasslands to the south
(Birks, 1976). At Wolsfeld and French lakes in southern
Minnesota grassland was present until ~5 Kya, then an oak
savanna developed and persisted until -300 years ago
when during the Little Ice Age Acer saccharum, Ostrya vir-
giniana, Tilia americana, and Ulmus arrived and formed
the present Big Woods of that region (Grimm, 1983). A de-
tailed history of Elk Lake, Minnesota, for the past 10 Ky is
presented by Bradbury and Dean (1993). In southern On-
tario oak savanna was also present after -6 Kya (Szeicz and
MacDonald, 1991). Prior to that study it was thought that
the oak savanna was a result of burning by Amerindians.
The history of Tsuga in the forest of the Great Lakes region
provides an example of factors that can delay the introduc-
tion of a species even though climates are suitable. In this
case the delay was due to the physical-biotic barrier pre-
sented by Lake Michigan surrounded along its southern
boundary by dry grassland (Davis et al., 1986).
Mean July temperatures at glacial maximum are esti-
mated at 10-12°C cooler than at present. The isotopic
composition of the waters of glacial Lake Agassiz in North
Dakota and southern Manitoba provides an estimate of av-
erage air temperature during the Late Wisconsin-Early
Holocene (11.7 Ka-9500 years B.P.) of-16°C compared to
the present 0°C (Remenda et al., 1994). Rainfall during cold
periods was 20% below the present in some areas (Bartlein
etal., 1984).
Model simulations (Kutzbach, 1987) generally parallel
292 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 8.14. Late Pleistocene lakes and marshes in the
Great Basin. Reprinted from Benson and Thompson (1987,
fig. 1, after Spaulding et al., 1983) with the permission of
the Geological Society of America.
results from the pollen and spore studies (Webb et al.,
1987). At 18 Kya temperatures in the central and eastern
sections of North America were an estimated 5°C colder
than at present, while in the southeast they were 1-2°C
colder. At 12 Kya the jet stream switched to a single flow as
the Laurentide ice sheet was reduced in area and elevation
(-50%). Temperatures in the north-central and northeast
regions warmed to 2-4°C cooler than at present and were
~1°C warmer in the southeast. At 9 Kya the ice sheet was
reduced to -800 m in elevation and July insolation was
-8% greater than now, producing the early stages of the cli-
matic optimum wherein temperatures across North Amer-
ica averaged 1-2°C warmer than at present. By 6 Kya the
Laurentide ice sheet had essentially disappeared and con-
ditions were at or near present values.
Southwest and West
One effect of the high-pressure anticyclone systems located
over the Laurentide and Cordilleran ice sheets during
glacial maxima was the displacement of low-pressure cy-
clone belts southward. During these times there was in-
creased precipitation and lower temperatures, and over
100 pluvial lakes formed in the southwest (Fig. 8.14). These
lakes muted seasonal temperature variations and served as
a positive feedback to increased precipitation. Tree lines
lowered by 500-1000 m as determined by Picea—Pinus ra-
tios from bogs compared to the ratios in surface samples
collected along altitudinal and vegetation gradients (e.g.,
Maher, 1963). Pine-juniper woodland moved into the
plains and basins, displacing desert shrubland and grass-
land, although recent studies indicate that Pinus may be
overrepresented at several sites and that desert grassland
was more prominent than thought earlier (Hall, 1985).
These conditions were reversed in each of the 18-20 inter-
glacial periods of the past 1.6 m.y. The fossil record from
the arid southwest shows that plant species there also re-
sponded to climatic change individually, forging combina-
tions that are now rare or absent (Thompson, 1988; Van
Devender et al., 1987). Physiographic diversity is greater
in the western cordilleran region of North America than in
the east or in the Plains. As a result, biotic history is more
reflective of localized basins, plateaus, slopes, and peaks.
Although correlated global and hemispheric events, such
as the Little Ice Age, the Younger Dryas, and the altither-
mal interval, are broadly evident throughout western
North America, the regional histories tend to be more in-
dividualized.
Pollen-bearing sediments are not extensive in this re-
gion, and diversity in assemblages representing low to
middle elevation habitats may be restricted to 40 or fewer
pollen types. Juniperus, Pinus, Gramineae, Artemisia, che-
noams, and Compositae often account for 90% or more of
the plant microfossils. Continuous and well-dated Holo-
cene sequences come mostly from montane bogs where
pollen of Abies, Picea, Pinus, and herbaceous pollen, usu-
ally undifferentiated to genus, are prominent. Conse-
quently, information on Quaternary biotas and environ-

Figure 8.15. Pollen diagram from Dead Man Lake, Chuska Mountains, New Mexico. Reprinted from
Hall (1985, fig. 7, after Wright et al., 1973) with the permission of the American Association of Strati-
graphic Palynologists Foundation.
ments must be assembled from fossil faunas, packrat mid-
dens, dendrochronology, megafossil assemblages, and
pollen from bogs, playa lakes, and cave and rock shelter
deposits (see papers in Jacobs et al., 1985). Also, the corre-
lation of vegetation change with climatic events is compli-
cated by topographic diversity, varied distances from
sources of oceanic moisture, and possible lag time or "veg-
etation inertia" (Cole, 1985; Spaulding, 1990a). The latter
concept is based on midden data from the Grand Canyon
region of northern Arizona. In the early stages of a new cli-
matic cycle the first organisms to disappear are those near
the limits of their ecological tolerance, which is recorded
in the diagrams as a decline in species diversity. However,
the dominants to some extent sustain their own microenvi-
ronment (sunlight, moisture, soil chemistry) and persist
longer into the cycle. Although some correspondence is
now emerging between high-resolution pollen diagrams
and the oxygen isotope record (Williams et al., 1993, fig.
8.22), if climatic changes are rapid, as in the Wisconsin-
Holocene transition, the plant record may not vary pre-
cisely in concert because of lag time. Another early source
of confusion in reconstructing regional climate and vegeta-
tional history, and comparing these with events in adjacent
regions, was pollen evidence suggesting that the interval
between -7.5 and 4 Kya years in the southwest was moist
Quaternary North American Vegetational History 293
(Martin, 1963a,b). Geomorphic and other data indicated it
was arid (the "long drought"; Antevs, 1938). The problem
was resolved with the discovery that the sections repre-
senting the altithermal at several sites (e.g., Whitewater
Draw, Arizona) had been removed by erosion.
Little is known about pre-Wisconsin vegetation in the
arid southwest, but in the far northwestern corner of Texas
the small Rita Blanca megafossil flora provides some insight
into Nebraskan age vegetation (Tucker, 1969). The promi-
nent member is Quercus that is similar to the Q. gambelii
that presently grows along the western forest-grassland bor-
der. Associates suggest higher water tables and more mesic
conditions than at present (Crataegus, Madura, Populus,
Ribes, Salix, Sapindus}. Drier elements include Amorpha
(mock lotus) and Baccharis (buckbrush); the pollen flora is
rich in Artemisia, other composites, and grasses.
For Wisconsin time more information is available from
the southwest. Many of the full-glacial records (24-14 Kya)
are rich in Artemisia and Pinus, have smaller amounts of
Abies and Picea (Fig. 8.15), and represent a cool sagebrush
steppe or sagebrush-pine woodland. The latter is an asso-
ciation of genera and a vegetation type that is not wide-
spread today. Alpine tundra expanded at the highest eleva-
tions as tree lines shifted downward. The transition from
glacial to postglacial conditions occurred between -14 and
294 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 8.16. Modern distribution (solid line) and Late Wis-
consin distribution (stippled) ofPinus edulis (pinon pine).
Arrows represent suggested migration routes of P. edulis
into its present range. Triangles are packrat midden sites.
Reprinted from Van Devender et al. (1984, fig. 3) with the
permission of Quaternary Research.
12 Kya in the southwest; after an Early Holocene continua-
tion of cool conditions, climates warmed in the Middle
Holocene and vegetation trended toward drier communi-
ties and an upward shift in ecotones.
Packrat middens from the Chihuahuan Desert of north-
eastern Mexico, Texas, and New Mexico show a sequence
from pinon pine (P. edulis)-jumper (/. scopulorum) wood-
land (mesic, Late Wisconsin), to oak-juniper woodland
(Early Holocene transition, -11 Kya), to the present desert
grassland-shrubland that developed between 8 and 4 Kya
(Van Devender, 1985,1990a; Van Devender et al., 1984). A
similar sequence was likely repeated throughout the Qua-
ternary as the low-pressure systems of the glacial intervals
contributed moisture and allowed deeper penetration of
cyclones from the Gulf of Mexico and the Pacific Ocean,
followed by the return of high-pressure systems in inter-
glacial times. Playa lakes were numerous and some of
these provided the basis for early pollen studies in the
southwest from the San Agustin Plains of New Mexico
(Clisby and Sears, 1956; see also Markgraf et al., 1984) and
the Wilcox Playa of Arizona (Martin, 1963c). Forests and
woodland grew 1000 m lower than at present and moun-
tain glaciers formed in the Sacramento Mountains of New
Mexico. These and other highlands likely served as refugia
for mesic elements that spread during glacial (pluvial)
times (Fig. 8.16). In each of the 18-20 principal inter-
glacials of the Quaternary, arid conditions returned and
vegetation similar to that of the present was established.
In the Trans Pecos region of west Texas the present veg-
etation is a desert scrubland of Flourensia, Larrea, grasses,
cacti, and composites. At higher elevations there is an
oak-juniper woodland. During the latest full glacial
(20-14 Kya) the vegetation was a pinon pine-juniper
woodland (Spaulding et al., 1983), reflecting cooler and
more moist conditions. The pine-spruce, ponderosa pine,
and pinon pine-juniper zones shifted downward and the
latter spread onto the lowland plains now occupied by
desert grassland and shrubland. By Holocene times condi-
tions at Bonfire Shelter in south-central Texas at the Mexi-
can border were warmer and drier than at present. Pollen
of grasses and composites increased after 10.5 Kya (Bryant
and Holloway, 1985). Later (8.7-6 Kya) pollen of the xero-
phytes Agave and Dasylirion increased, corresponding to
the warm-dry climate of the climatic optimum (Hall, 1985).
Rainfall continued to decrease until -2.5 Kya, based on
faunal records from the Edwards Plateau region (disap-
pearance of high moisture-requiring taxa Pipistrellus sub-
flavus, pipistrelle bat; Microtus pinetorum, woodland vole;
Toomey et al., 1993). Arid late-glacial and interglacial con-
ditions are also suggested by landscape morphology and
pollen assemblages from the San Juan Basin of northwest-
ern New Mexico (Hall, 1977,1985,1990) and by low water
tables in the southern High Plains of Texas (Meltzer, 1991).
After -2.5 Kya slightly more mesic conditions returned to
the Edwards Plateau region, and essentially modern biotas
and environments have prevailed.

To the northwest along the Texas-New Mexico border,
early pollen studies were made at several sites on the Llano
Estacado (southern High Plains; Hafsten, 1964). The pres-
ent rainfall there is 400—500 mm, average January temper-
ature is ~5°C, and average July temperature is ~27°C. The
vegetation is a shrubland (juniper, oak)-desert grassland
(Bouteloua, grama grass; Buchloe, buffalo grass; cheno-
pods; composites; Artemisia, sagebrush; Ephedra; Kiichler,
1964). At about 33.5 Kya the vegetation was an open wood-
land of pine and spruce, summer temperatures were an es-
timated ~5°C cooler than at present, and moisture was
more plentiful. The principal fluctuations in vegetation in-
volved decreases in trees with more Artemisia (warmer
and drier), then a return to pine and spruce parkland. At
the Wisconsin glacial maximum at 22.5-14 Kya, corre-
sponding to an 8-10°C cooler interval, more closed wood-
lands of pine and spruce prevailed. Near the end of the
Wisconsin stage the Artemisia steppe conditions of the
present were attained. It is assumed that a similar history
of woodland (with boreal and western montane conifers),
to less dense parkland, to open shrubland with abundant
grasses characterized the region from the Nebraskan glacial
stage throughout the Quaternary. In the nearby Hueco
Mountains (El Paso and Hudspeth Counties, Texas) packrat
middens preserve a pinon pine-jumper-oak woodland at
~13.5-12 Kya, changing to drier oak-juniper woodland by
-9.4 Kya, to desert grassland (8.9-4 Kya), to modern Chi-
huahuan Desert scrub after 4 Kya (Van Devender and Ris-
kind, 1979; Van Devender et al., 1987). At Chaco Canyon in
northwestern New Mexico (Betancourt and Van Devender,
1981), pollen evidence suggests less pinon pine-juniper
woodland (midden data) and more shrubby grassland (Hall-
1981, 1982, 1985). If the pine and juniper were localized
stands in canyons, they may be overrepresented in the
middens; and the vegetation may have been mostly an arid
shrub grassland for the past 7 Kya (Hall, 1977).
The Puerto Blanco Mountains of southwestern Arizona
are located in one of the most arid regions of North Amer-
ica (Organ Pipe Cactus National Monument, Sonoran
Desert). A Holocene history has been reconstructed from
middens (Van Devender, 1987). The sequence is from ju-
niper-joshua tree woodland (-14 Kya) to desert shrubland
with Carnegiea gigantea (saguaro) and Encelia farinosa
(brittle bush) as dominants and Acacia greggii, Prosopis ve-
lutina, and Cercidium floridum as associates (-10.5 Kya).
There followed Sonoran Desert vegetation with Sapium
biloculare (Mexican jumping bean), Olneya tesota (iron-
wood), and Stenocereus thurberi (organ pipe cactus) at -4
Kya. In the Waterman Mountains just to the east in south-
ern Arizona, the Middle to Late Wisconsin vegetation was
pinon pine-juniper woodland with Artemisia (sagebrush),
Vauquelinia californica (Arizona rosewood), and Yucca
(Joshua tree) at 22.5-11.5 Kya, followed by communities
similar to those recorded at Puerto Blanco (Anderson and
Van Devender, 1991; Van Devender, 1990b). The Clovis
complex at Murray Hills, Arizona, is the oldest docu-
Quaternary North American Vegetational History 295
mented Paleolndian tradition in the southwest. It includes
remains of browsing mammoth and grazing bison and
dates from -11.2 to 10.9 Kya (Hoffecker et al., 1993).
Middens from the Scodie Mountains in the western
Mojave Desert are from 13,330 to 12,870 years old. The site
is at 1215 m elevation and the present vegetation is a Lar-
rea tridentata (creosote bush)-Ambrosia dumosa (bur sage)
desert. In the uplands there is a Joshua tree (Yucca brevifo-
lia) desert at 1300-1900 m and a pinon pine (P. mono-
phylla)-oak (Q. chrysolepis) woodland at 1900-2200 m.
At -13 Kya Death Valley (elevation -86 m) in the Mojave
Desert was a ~173-m deep lake (90 m from 10 to 35 Kya; Li
et al., 1996). In the Late glacial there was a pinon pine-
juniper (/. californica] woodland (Thompson, 1990). Lower
ecotones, plants no longer in the immediate vicinity
(54.5%; e.g., Ceanothus greggii), and modern analogs else-
where suggest a January mean temperature of -0.75°C
(presently 2.5°C), a July mean of 21.3°C (25.3°C), and pre-
cipitation of 218 mm/year (168 mm/year; McCarten and
Van Devender, 1988). The MAT at glacial maximum (-18
Kya) has been estimated to have been ~6°C colder than at
present. A second set of middens from the McCollough
Range in the southeastern Mojave Desert of Nevada are
6800-5060 and 1250 years old (Spaulding, 1991). The pres-
ent vegetation is dominated by Larrea tridentata and Aca-
cia greggii, with Ephedra torreyana, Chrysothamnus tereti-
folius (rabbit brush), Encelia cf. virginensis (brittlebush),
Krameria parvifolia, Opuntia basilaris, Peucephyllum
schottii (desert spruce), and Viguiera reticulata. Coopera-
tive Holocene Mapping Project (COHMAP) reconstructions
(COHMAP Members, 1988) depict Middle Holocene arid-
ity (climatic optimum) for the Pacific Northwest and the
Great Basin, which may have extended into the Mojave
Desert and parts of the Sonoran Desert (Hall, 1985). The
6800-5060 year old McCollough Range samples (Middle
Holocene) do reflect drier conditions. The most xeric mem-
bers (e.g., Larrea, Peucephyllum] are consistently more
abundant than in the Late Holocene or modern assemblages,
while more mesic species are less abundant (Chrysotham-
nus, Encelia). Studies of middens in the Mojave Desert gen-
erally indicate that in the last glaciation juniper and pinon
pine woodlands were widespread (Spaulding, 1990b). The
woodlands gave way to desert scrub in the latest Wisconsin
and Early Holocene, with possible reversals to brief moist
conditions at -10 and 8 Kya. Early Wisconsin (pre-24 Kya)
pollen sequences from Tule Springs, Nevada (Mehringer,
1965,1967), and elsewhere show that Artemisia was in the
lowlands and that zones of Abies, Picea, and Pinus were
lower by 900-1400 m. The most recent version of the Mo-
jave Desert vegetation was established by 7 Kya, although
the meager records from the area do not preclude later fluc-
tuations in vegetation. In the White Mountains along the
California-Nevada border, southeast of Yosemite National
Park, the hydrogen isotope composition of bristlecone pine
tree rings provides a temperature chronology for the past 8
Ky (Feng and Epstein, 1994). The postglacial climatic opti-
296 Late Cretaceous and Cenozoic History of North American Vegetation
mum is recorded at -6.8 Kya followed by gradual cooling
then rapid cooling at -1700 A.D. (Little Ice Age). Both events
are also reflected in tree-ring analyses in the southern
Sierra Nevada (Scuderi, 1993).
A pollen assemblage from the Santa Barbara Basin in
southern coastal California shows a change from upland
coniferous forest (Pinus) to lower montane and lowland
forest (Quercus with Artemisia) between -12 and 7.8 Kya,
then to modern chaparral (Adenostoma, Cercocarpus,
Rhamnus, Rhus, Ceanothus) and coastal sage scrub after
5.7 Kya (Artemisia, Erigonum, Salvia; Heusser, 1978). Rel-
ative frequency and pollen influx diagrams reflect the re-
duction in vegetation density. The inferred climate is cool
(1-2°C cooler) and wet (150-250 mm more rainfall) with a
300-m lowering of tree line in the late glacial, trending to
warm and dry by the climatic optimum. Treeline studies in
the White Mountains of California suggest temperatures
~2°C lower at 6 Kya and general estimates of temperature
for the U.S. Pacific Northwest are 6°C in the early post-
glacial and 4°C in the late postglacial. Middle Holocene
warmth is evident in profiles from the Steens Mountains of
southern Oregon (Mehringer, 1985).
Pollen and plant megafossil profiles from three high-
altitude lakes in the central Sierra Nevada (-2400-3000
m), near the western edge of Yosemite National Park, record
vegetation for the past 12.5 Ky (Anderson, 1990). Prior to
10 Kya trees were sparse to absent, probably restricted by
poor soils following deglaciation. Between 10 and 6 Kya
Pinus flexilis (limber pine), P. monticola (western white
pine), and P. murrayana (lodge-pole pine) become estab-
lished at various sites. Montane chaparral shrubs of Arc-
tostaphylos sp., Artemisia sp., Cercocarpus betuloides
(mountain mahogany), and Chrysolepis sempervirens (bush
chinquapin) were abundant, suggesting the forest was
open. Abies magnifica (red fir), Pinus albicaulis (whitebark
pine), and Tsuga mertensiana (mountain hemlock) were
confined to mesic habitats. The inferred climate during
this interval was warm and dry. After -6 Kya precipitation
increased, which is shown by a greater abundance of the
mesic Abies magnifica, Pinus murrayana, and Tsuga
mertensiana and fewer dry shrubs and herbs. Cooling is
evident by 3- 2.5 Kya with downward shift in the upper al-
titudinal limits of Abies magnifica and Tsuga mertensiana
and in the lower limits of Pinus albicaulis. A similar low-
ering of ecotones occurred elsewhere in the Sierra Nevada
region (Koehler and Anderson, 1994): the montane conifer
forest was lower by 300-700 m in Kings Canyon (Cole,
1983), and Artemisia and Sarcobatus move into the pres-
ent grasslands of the San Joaquin Valley (Atwater et al.,
1986).
Long pollen profiles from Clear Lake near the southern
terminus of the northern California Coast Ranges provide a
record of climates and vegetation at an elevation of 404 m
(Adam, 1988). The cores cover the interval from the Late
Illinoian to the present. The modern vegetation is mostly
an oak woodland (Q. douglasii, blue oak) with digger pine
(Pinus sabiniana). Other smaller and/or more distant for-
mations include mixed hardwood forest, chaparral, and
Coast Range montane coniferous forest. Despite the lengthy
time span represented by the core, the vegetation and cli-
matic changes throughout are reflected by just two com-
ponents. There are numerous rapid shifts between asso-
ciations dominated by oak (warm and dry) and those
dominated by pine and TCT-type pollen (cool and moist).
An interesting feature is that the patterns correlate well
with those for the Eemian and Wiirm (Weichsel-Deven-
sian) of Europe. The implication is that if cores are long,
continuous, well-dated, and sufficiently near the ocean,
they may preserve responses to Milankovitch (long term)
and thermohaline—ocean circulation (shorter term) forcing.
Additional evidence is accumulating that the abrupt
climatic changes that characterized the North Atlantic also
affected the North Pacific region. A 196-m core from the
Santa Barbara Basin records rapid shifts in climate during
the past 20 Ky that parallel those from the Greenland ice
cores and adjacent marine sediments (Kennett and Ingram,
1995). Another core from the subarctic Pacific Ocean ex-
tends the pattern back to 95 Kya (Kotilainen and Shackle-
ton, 1995), as do results from the eastern North Pacific
Ocean (Thunell and Mortyn, 1995). Hydrological cycles of
the Owens Basin in the Great Basin of northwestern Cali-
fornia during the last glacial termination parallel climatic
changes in the North Atlantic (Benson et al., 1997). Corre-
lation between cosmogenic chlorine-36 dates on moranes
in the Sierra Nevada and radiocarbon chronologies from
Owens Lake reveal that glacial advances correspond to
Heinrich events 5,3,2, and 1 (Phillips et al., 1996). Diatom
sequences and fluctuations in percentages of Juniperus
pollen reflect cooler and more moist climates during gla-
cial intervals by displacement of storm tracts from the
Alutian low pressure system (Bradbury, 1997a,b; Smith et
al., 1997). In sediments from the Olympic Peninsula of
Washington, and continuing north to the Alaska Panhan-
dle, there are peaks of Tsuga mertensiana pollen, reversals
from forested to nonforest vegetation, and other evidence
for a cold interval between 11 and 10 Kya suggesting a
Younger Dryas climatic reversal (Mathewes, 1993). These
new data provide a context for interpreting terrestrial envi-
ronmental history of the Pacific Northwest comparable to
that provided by the Greenland data for the North Atlantic.
They are also of theoretical importance because they show
that climate forcing mechanisms in the Quaternary may
have been circumpolar in extent, which would require that
developing models account for such correlated hemi-
spheric events. One implication is that freshwater influx
into the North Pacific Ocean from the Cordilleran ice sheet
may have interfered with some degree of thermohaline cir-
culation.
Along the coast an indication of Late Pleistocene envi-
ronments is provided by the remarkable fossil fauna at
Rancho La Brea (the La Brea tar pits; Stock, 1992). About
600 different kinds of organisms are known and include

bats, imperial mammoth, mastodon, saber-toothed cat
(abundant), puma, lynx, dire wolf (abundant), coyote
(abundant), bear (short-faced, black, grizzly), weasel, bad-
ger, skunk, rodents (gopher, mice, kangaroo rat, voles,
ground squirrel, rat, chipmunk, jackrabbit, brush rabbit,
cottontail), antelope, pronghorns, peccary, bison, camel,
tapir, horse, deer, ground sloth, and giant ground sloth. The
plants have not been studied but include a chaparral asso-
ciation of Adenostoma (chamise), Arctostaphylos (man-
zanita), Ceanothus (buckbrush, California lilac), Juglans,
Quercus, Sambucus (elderberry), and others.
In the Great Basin region, Great Salt Lake in Utah has a
surface area of 6200 km2. Its predecessor glacial Lake Bon-
neville at 18 Kya had a surface area of 47,800 km2
(Hostetler et al., 1994) to 51,300 km2 (Benson and Thomp-
son, 1987), a volume of 9500 km3, and was -300 m deeper
than at present. Late Pleistocene precipitation is estimated
at 1.6-1.9 m/year (present 0.4-1.5 m/year) or 33% greater
than at present (Lemons et al., 1996). Later glacial Lake La-
hontan in western Nevada reached a size of 14,700 km2
(remnants presently total 2500 km 2 ; Benson and Thomp-
son, 1987) during the 1400-year period between Heinrich
event 1 at 14 Kya and the resumption of thermohaline cir-
culation at 12.7 Kya. Zones of permafrost in the uplands of
Idaho, Montana, Wyoming, and Colorado were lowered by
1000 m and MAT is estimated at 9-ll°C below that of the
present (Pewe, 1983). In the southern Great Basin MAT is
Quaternary North American Vegetational History 297
Figure 8.1 7. General distribution of common species of
Juniperus in the western United States. JUOS, /. os-
teosperma (Utah juniper); JUSC, /. scopulorum (Rocky
Mountain juniper); JUMO, /. monosperma (one-seeded
juniper); JUDE, /. deppeana (alligator juniper); OC, /. oc-
cidentalis (western juniper var. occidentalis); JUOC-AU,
/. occidentalis (western juniper var. australis); JUPI, /.
pinchotii (Pinchot's juniper); JUER, /. erythrocarpa (red-
berry juniper). Reprinted from Miller and Wigand (1994,
fig. 2, based on Critchfield and Little, 1966) with the
permission of Richard Miller, Peter Wigand, and the
American Institute of Biological Sciences.
estimated to be 6°C cooler and precipitation 30-40%
greater (Spaulding, 1990b). The Middle Holocene period of
aridity between -7.5 and 4.5 Kya, as proposed early by An-
tevs (1938), has generally been confirmed, although it may
have been expressed at slightly different times in various
parts of the west.
The montane coniferous forest shifted downward by
-600 m and its range expanded during the cool moist in-
tervals of the glacials and Early Holocene (Picea, Pinus
flexilis; Abies locally). The lower limit of the forest was at
-1000 m, compared to the present 1500-2000 m limit.
Below the montane coniferous forest, the mountain slopes
between -1500 and 2250 m are presently occupied by
pinon-juniper woodland (Fig. 8.17). Farther down on the
lower slopes and basin floor there is a woodland (steppe) of
Artemisia. It is obvious from Fig. 8.17 that the regional veg-
etation is a complex mosaic of communities and that even
slight environmental change would have produced consid-
erable response from biotas arranged along the steep altitu-
dinal gradients. Changes did occur throughout the Quater-
nary and they are particularly evident in the turbulent
transitions between glacial and interglacial stages. During
cool moist glacial intervals the drier habitat species of Ju-
niperus were 500-600 km further south and up to 1500 m
lower in elevation (Miller and Wigand, 1994). Mesic
species (e.g., /. osteosperma, Utah juniper) extended into
the Mojave and Sonoran deserts (Thompson, 1990). In the
298 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 8.18. Diagram of relative
abundance ofjuniperus pollen,
and ratios of Gramineae to
Artemisia and charcoal to
pollen, Diamond Pond, south-
eastern Oregon. Reprinted from
Mehringer and Wigand (1990)
with the permission of the Uni-
versity of Arizona Press.
Middle Holocene climatic optimum (-7-5.4 Kya) there
was widespread shrubland (steppe) of Artemisia, Atriplex
confertifolia (shadscale), and Sarcobatus (greasewood) in
the lowlands, followed after -4.5 Kya by slight expansion
of the coniferous forest at high altitudes and more mesic
species of juniper (downslope -150 m; Mehringer, 1985;
Figs. 8.18-8.20). Similar fluctuations probably occurred in
earlier glacial times (Coplen et al., 1994). Throughout the
past 1.6 m.y., however, the seasonally cold dry shrub-
land-steppe has been the dominant vegetation at low to
midelevations of the Great Basin; at times it was even more
extensive than at present.
In Grand Teton National Park, Wyoming, the vegetation
near the end of the Pinedale Glaciation at -14-11 Kya was
alpine meadow with Betula and Juniperus (Whitlock,
1993). Treeline was lower by ~600 m and temperatures
were an estimated 5-6°C cooler. Temperatures and winter
precipitation increased between 11.5 and 10.5 Kya, allow-
ing Picea, Abies, then Pinus cf. albicaulis to invade. By
-10.5 Kya modern subalpine forest was established. At Yel-
lowstone National Park, Wyoming, highland sites prior to
10 Kya were treeless (Whitlock, 1993); at slightly lower al-
titudes pollen from a pond core included funiperus, Picea,
Cyperaceae, Gramineae, Artemisia, Betula, and Salix, indi-
cating a treeline lower by -500 m (Waddington and Wright,
1974).
Glacial history is known in considerable detail for Glac-
ier National Park in Montana (Carrara, 1989); along with
dated ash sequences, a high-resolution context is available
for reconstructing the vegetational history (Barnosky et al.,
1987a; Mack et al., 1983). Highland areas were mostly
deglaciated by 11.2-11.4 Kya and long before that in the
adjacent plains, as shown by fossils below the Glacier Peak
(11.2 Kya) and early Mount St. Helens (11.5 Kya) ash lay-
ers. Pollen, wood, needles, and insects below the Mount St.
Helens ash show that treeline was ~500-700 m lower than
at present. At -11.2 Kya in the Kootenai River Valley, 150
km west of Glacier National Park, basal sediments from a
lake at 1270-m elevation contain pollen of pine, spruce, fir,
alder, and willow, suggesting a cool moist climate (Mack et
al., 1983). At Sheep Mountain bog 150 km south of the
park, douglas fir has occupied the site since at least 10 Kya
(Mehringer, 1985). Reforestation in the highlands at Glacier
National Park began -10 Kya as an open conifer forest of
Pinus contorta (lodgepole pine) with spruce (pollen) and
spruce or larch (wood). This type of vegetation is charac-
teristic of the present alpine tundra-subalpine forest
boundary at treeline. By 10 Kya the glaciers were occupy-
ing the cirques and depressions of the present; similar
chronologies are known from the North Cascades, Banff,
Jasper, and Yoho National Parks (pine-fir forests at 10 Kya).
At -9.5 Kya the treeline was near its present position in the
Front Range of Colorado (Benedict, 1973) and -80 m higher
in the San Juan Mountains of southwest Colorado (Carrara
et al., 1984). At lower elevations the sequence has been from
mostly pine-dominated (warm) to Artemisia- dominated
(colder) steppe tundralike vegetation. Longer records from
Yellowstone National Park document that similar shifts in
treeline, involving essentially modern regional communi-
ties, probably occurred throughout the Quaternary (Fig.
8.21).
After the late-glacial to Early Holocene cool period, cli-
 Quaternary North American Vegetational History 299

Figure 8.19. Pollen diagram from Swan Lake, Bannock County, Idaho. Reprinted from
Mehringer (1985, fig. 5, and references cited therein) with the permission of the Ameri-
can Association of Stratigraphic Palynologists Foundation.

mates in the northern Rocky Mountains warmed to the lein (1997). The Quaternary history of coastal communities
Middle Holocene climatic optimum. MAT at 5900-m eleva- from southern Oregon to British Columbia during glacial
tion in Jasper National Park was ~1.9°C warmer than at times is generally one of lowered forest ecotones, cold dry
present (Osborn, 1982). In the mid-1700s there was a briefsteppe in inland areas, and subalpine-type forests along
period of glacial advance correlated with the Little Ice Age,
the coast. The downward shifts did not involve entire veg-
and records of fir pollen indicate that treeline reached its
etation zones, only differential movement of species ac-
lowest altitudinal limit in the last 500 years (Kearney and
cording to their individual ecological requirements. Thus,
Luckman, 1983). mixtures of montane and lowland elements developed that
do not occur today. This is consistent with the individual
responses of various taxa to Quaternary climatic change
Northwest
described for the eastern deciduous forest and the arid
The maximum extension of continental ice in the Pacific southwest (e.g., sagebrush-pine). Further north in Alaska,
Northwest was near Olympia, Washington, in the Puget the Wisconsin and earlier glacial-stage vegetation was
Lowland, and in the northern Columbia Basin at -14.5 Kya. mostly tundra of modern aspect.
Early pollen studies in this region were made by Hansen When this paleoecological information is used as vali-
(1939 et seq., see summary, 1947), Heusser (1952, 1977 et dation data for CCM simulations (Kutzbach, 1987), the
seq.), and Mathewes (1973, 1979), and more recently by model results are generally supported. At glacial maxima
Barrie et al. (1994), Mann and Hamilton (1995), Mathewes the splitting of the jet stream by the Cordilleran and Lau-
(1985,1991), Mathewes and Clague (1994), McLachlan and rentide ice sheets brought the southern arm across latitude
Bmbaker (1995), Mock and Bartlein (1995), Thompson et -45° N (Oregon-Washington border). An anticyclone sys-
al. (1993), Whitlock (1992, 1993), and Whitlock and Bart- tem developed between this latitude and the glacial margin
Figure 8.20. Summary of vege-
tation and inferred climates
for northeastern Washington
and northern Idaho since
the Wisconsin glaciation.
Reprinted from Mack et al.
(1978, fig. 3) with the per-
mission of The University of
Chicago Press.

Figure 8.21. Estimated vegetational changes for the past 135 Kya, Yel-
lowstone National Park. Reprinted from Baker and Wain (1985, fig. 3,
and references cited therein) with the permission of the American Asso-
ciation of Stratigraphic Palynologists Foundation.
 Quaternary North American Vegetational History 301

Figure 8.22. Correlation diagram of pollen zones from

Quaternary sections in Washington and Oregon. Triangles
indicate dated levels. Reprinted from Heusser (1985, fig. 11)
with the permission of the American Association of Strati-
graphic Palynologists Foundation.

to the north, resulting in increased easterly winds off the
Laurentide and Cordilleran ice sheets, cooler temperatures,
and drier conditions east of the Cascade Mountains. The
simulation is paralleled in the fossil record by steppe and
tundra characteristic of cold dry climates in rainshadow
areas. By 12 Kya July temperatures were warming as indi-
cated by model simulations (Kutzbach, 1987) and valida-
tion data (Barnosky et al., 1987b). This low precipitation
signal is in contrast to pluvial conditions in the southwest
during glacial maxima.
Only maritime areas to the west of the Coast Ranges re-
mained warm and wet. During interglacial times, the Pa-
cific coast coniferous forest (Pinus contorta, lodgepole
pine) and boreal forest (Picea sitchensis, sitka spruce) mi-
grated from refugia (e.g., western Olympic Peninsula, Queen
Charlotte Islands, interior Alaska) and established a distri-
bution and composition similar to that of the present
(Heusser, 1985); Fig. 8.22). Pollen diagrams from Hoh-
Kalaloch, Washington, show tundra or park tundra at gla-
cial maxima and elsewhere the later development of
302 Late Cretaceous and Cenozoic History of North American Vegetation
spruce parkland, pine woodland, and forest communities
after 13 Kya. At 6.3 Kya the vegetation at Devils Lake in
coastal Oregon included Picea sitchensis, Alnus, Lysichi-
ton (Araceae), and ferns, with later introduction of Tsuga
heterophylla (western hemlock), Thuja plicata (western
red cedar), Pseudotsuga menziesii (douglas fir), Pinus, and
Alnus. At Carp Lake in the Columbia Basin of south-central
Washington the full-glacial vegetation was a cool, dry,
nonarboreal steppe of Artemisia and grasses with alpine el-
ements in the highlands (Barnosky, 1985; Whitlock and
Bartlein, 1997). MAT is estimated at ~10°C (presently
~15°C) and precipitation at 1300 mm along coastal Oregon
during the Wisconsin glacial maximum (presently ~2600
mm). At 8 Kya temperatures increased to ~13°C and pre-
cipitation to -2400 mm, representing the period of maxi-
mum warmth in the region. After that time the climate be-
came slightly warmer and wetter.
At glacial maximum ~14,400 years B.P. in coastal
British Columbia, this cold interval corresponds to Hein-
rich event Hi as represented by coarse sediments in an
otherwise fine sediment sequence from the continental
slope (Blaise et al., 1990; Mathewes, 1996). As noted in the
discussions of Clear Lake and glacial chronologies in the
Sierra Nevada, detailed analyses of pollen cores near
coastal and other areas in the west are revealing fluctua-
tions in closer accord with orbital forcing, ocean circula-
tion, and disruption of some level of thermohaline circula-
tion. Although such correlations are not yet numerous and
exact, there is growing evidence to suggest that Quaternary
climatic histories in the North Pacific and the North At-
lantic were more similar than previously thought (see
Mathewes, 1993; Mikolajewicz et al., 1997). It should be
noted, however, that lower salt content of northern Pacific
Ocean surface waters prevents the same degree of sinking
as in the northern Atlantic Ocean. After glacial retreat, the
forest sequence in the lowlands of British Columbia during
the Early Holocene was Pinus contorta (cool, probably
dry); Abies, Picea, Pinus, and Tsuga (cool and moist); and
modern vegetation at 4-2 Kya. At the same time the high-
lands of the southern interior supported Artemisia and
grasses. Other fluctuations in various parts of British Co-
lumbia are detailed by Hebda (1995).
The region of eastern Washington (Johnson et al., 1994)
experienced catastrophic flooding between ~15.3 and 12
Kya that formed the scablands (Chapter 2, Missoula
Floods). At nearly the same time (11,250 years B.P.) two
falls of the Glacier Peak ash, about 25 years apart, blan-
keted parts of the Pacific Northwest to a depth of ~6 m
(Mehringer et al., 1977). The ash is an important marker for
dating events in the region, and a layer is present in the
flood sediments. Undoubtedly both the floods and ash falls
affected the vegetation, but their impact is difficult to iden-
tify from pollen and spore diagrams presently available. In
the case of the ash fall, the recovery time from colonizing
vegetation to woodland (steppe) and forest was brief and is
not evident on pollen diagrams from adjacent regions
when standard sampling intervals are utilized. Scouring by
the floods removed all sediments, so records in the imme-
diate vicinity begin after the event. Profiles from adjacent
regions might reflect some change in the vegetation due to
catastrophy, but these would likely be subtle and blurred
by local pollen and spore input. Fine-resolution sampling
and appropriate statistical tests (e.g., analysis of variance)
may eventually allow the effects of the ash fall and flood-
ing to be identified and isolated from those caused by fluc-
tuating climates at the Pleistocene-Holocene transition.
In interior and western Canada a rich array of localities
are available for interpreting Quaternary vegetational his-
tory and paleoenvironments (e.g., see Ritchie, 1984, 1985,
table 1 and fig. 2). A site along the Roaring River in Mani-
toba preserves a pollen record for the Middle Pleistocene.
The sequence is boreal forest (Picea; cool climate), grass-
lands (Gramineae, Artemisia, chenoams), and oak savanna
(Quercus; warm), and a return back to boreal forest
(Klassen et al., 1967). A site on Baffin Island (Miller et al.,
1977; Terasmae et al., 1966) is tentatively assigned to Iso-
tope Stage 5e (Sangamon interglacial) and the vegetation is
a low Arctic shrub tundra (Alnus, Betula, Ericaceae, Salix).
Megafossils indicate that the birches were of the dwarf-
shrub type (B. glandulosa, B. nana). Mean July tempera-
ture is estimated at -3° C warmer than at present. In east-
central Alberta, Canada, the sequence is birch-spruce-
dominated boreal forest (-11.4 Kya), giving way to open
parkland-grassland by -10 Kya in response to Holocene
warmth (Hickman and Schweger, 1996).
Late glacial and Holocene records from British Colum-
bia, Alberta, and the Yukon Territory were summarized by
Ritchie (1985) and show a general sequence from late
glacial to Early Holocene of tundra giving rise to boreal for-
est (Picea) and a Middle Holocene period of warmth char-
acterized by increased Artemisia (southern Canada), grasses
(central), or mixed coniferous forest (northern).
Vegetational history in the western Arctic region is re-
viewed by Lamb and Edwards (1988). In southeastern
Alaska winds off the Gulf of Alaska bring moisture inland,
and the area was extensively glaciated during the Quater-
nary. Hence, most pollen records begin in the Late Pleis-
tocene and Holocene. In the Aleutian Islands tundra vege-
tation is evident throughout sections that date from the last
glacial retreat at 12-10 Kya (Heusser, 1985,1990; Fig. 8.23).
Inland, an early sedge tundra (grasses, willow, sage; 25-14
Kya) is supplemented by dwarf birch (14-12 Kya), fol-
lowed by an Alnus shrub tundra with aspen and balsam
poplar between -10 and 7 Kya. Spruce first became estab-
lished at 9.5 Kya, western hemlock came in at -3.5 Kya,
and Tsuga mertensiana (mountain hemlock) arrived at 3
Kya. There is evidence that the cold Younger Dryas interval
(11-10 Kya), well documented in the North Atlantic and
western Europe, may be evident in the Alaskan sequences.
Between 10.8 and 9.8 Kya forest parkland was replaced by
shrub tundra followed by spruce and hemlock. If the tun-
dra developed in response to Younger Dryas cooling, it

Figure 8.23. Correlation diagram of pollen assemblage zones of Quaternary sections in Alaska. Reprinted from Heusser
(1985, fig. 9) with the permission of the American Association of Stratigraphic Palynologists Foundation.
would mean that this event, previously ascribed to changes
in North Atlantic ocean circulation, may have been more
global in scope (Engstrom et al., 1990).
Winds and moisture were prevented from penetrating
into the interior by the Alaska Range, and much of this re-
gion remained ice free during the Quaternary. Thus, it
served as an important refugium for biotas that spread out-
ward during the warm interglacials onto newly exposed
surfaces (Ager and Brubaker, 1985; Edwards and Barker,
1994, northeastern Alaska). Examples of the interactions
between herb tundra, shrub tundra, and boreal coniferous
forest are shown in the composite diagram for Tanana Val-
ley Lakes near Fairbanks (Fig. 8.24), and the Late Wisconsin
and Holocene history for south-central and interior Alaska
is summarized in Fig. 8.25. In the central Mackenzie Moun-
tains of Canada an Artemisia and Salix herb tundra was
present from at least 12-10.2 Kya, followed by a shrub tun-
dra of Betula glandulosa and Populus balsamifera. Picea
arrived by 8.5 Kya and, with the expansion of Alnus, the
present vegetation was essentially established by 6 Kya.
The presence of Artemisia pollen opens the possibility
that regions within the Arctic Circle may have supported
sagebrush steppe during glacial maxima (most recently be-
tween 22 and 14 Kya). This interpretation is consistent
with abundant bones of mammoths, horses, bison, and
other ungulates in Beringian sediments—the mammoth
steppe. Pollen analysis suggests mostly tundra, and the rel-
ative prominence of the two ecosystems is unsettled. New
ocean cores from submerged portions of Beringia in the
Bering and Chukchi Seas have yielded meager amounts of
Artemisia pollen, and the profiles are interpreted as reflect-
ing tundra (Elias et al., 1996). The attractiveness of a steppe
is that it is better suited to support large herds of ungulates
Quaternary North American Vegetational History 303
and, in turn, human hunters following their migrations
into North America. As noted by Colinvaux (1996), a spir-
ited response from the big-mammal community doubtless
will be forthcoming.
Early pollen studies on Alaskan Quaternary environ-
ments were made by Livingstone (1955, 1957) and Colin-
vaux (1963, 1964a,b). This work is important because it
documented that the Arctic was cold during times of
glaciation, in contrast to a theory proposed by Ewing and
Donn (1956) that the region must have been warm to pro-
vide the moisture necessary for the growth of glaciers. Ice-
wedge evidence indicates that the MAT was -6°C during
the Wisconsin and similar conditions probably prevailed
during previous glacial intervals. As noted in Chapter 2
(Orogeny), model simulations show warm winters in
unglaciated interior Alaska while the ground evidence
shows that the vegetation was tundra and the climate cold
and dry. Approximately all of northwestern Canada and
82% of Alaska is now underlain by permafrost (Pewe,
1983), and on the North Slope in the vicinity of Barrow
and Prudhoe Bay it is presently 400 m thick (Walker, 1973).
VEGETATION SUMMARY
In eastern North America the deciduous forest at the for-
mation level was modernized in the Pliocene with the dis-
appearance of present-day Asian genera. Beginning -2.4
Ma Tertiary predecessors of the Appalachian montane
coniferous forest coalesced and spread from high-altitude
sites. Picea was present in Jackson (Late Eocene) time,
Tsuga is known from the Middle Miocene, and Abies ap-
pears in the Pliocene in the southern Appalachian region.
304 Late Cretaceous and Cenozoic History of North American Vegetation

Figure 8.24. Composite pollen diagram from several sites in the Tanana Valley, Alaska.
Reprinted from Ager and Brubaker (1985, fig. 8) with the permission of the American As-
sociation of Stratigraphic Palynologists Foundation.

This association was prominent during the subsequent
cool to cold glacial intervals when present-day boreal ele-
ments were also periodically introduced and eliminated
from the eastern forest (e.g., Pinus banksiand). A form of
alpine tundra grew at the highest elevations in the central
and northern Appalachian Mountains and perhaps at more
restricted sites to the south. The cold climates and Ap-
palachian coniferous forest are now recognized as typical
for -90% of Quaternary time. The deciduous forest mostly
occupied refugia such as those identified from southwest-
ern Tennessee, south-central Alabama, and northern Flor-
ida. During the brief warm intervals it spread from these
refugia to form the extensive, and exceptional, modern for-
mation while the coniferous forest moved into the high-
lands. This scenario was probably repeated in each of the
18-20 cold-warm cycles of the Quaternary and reveals
the temporal and dynamic nature of the Appalachian mon-
tane coniferous forest and the deciduous forest.
Much of Florida and the southern United States coastal
plain was inundated during interglacial times, and the
flood-plain forest association expanded. With cooling cli-
mates, glaciers advanced, sea levels lowered, and the
coastal plains were vegetated by forests capable of growing
under cold conditions and in the physiologically arid deep
sand. In Florida the plant cover alternated between open
pine—oak forest with understory grasses in the warm inter-
vals and pine forest during colder times. These changes
correlate with Heinrich events and suggest that alterations
in ocean circulation were a primary forcing mechanism. To
the north in the Carolina Bays region it was Picea rather
than extensive deciduous elements that grew with Pinus
along the coastal plain during the glacial intervals. The lat-
est version of the pine woods association begins at ~5 Kya.
In the far southwestern portion of the deciduous forest re-
gion (to central Texas), deciduous elements with some
spruce moved into mesic sites from the east and north dur-
ing cool periods; drier areas supported oak-hickory savan-
nas. The latter expanded in the warmer interglacial cli-
mates, and the pattern repeated throughout the Quaternary.
In the northwestern part of the deciduous forest the inter-
play was between forest, grassland, and spruce (with ash)
savanna.
To the northeast, a four-part chronology describes the
vegetation from the last glaciation to the present; by infer-
 Quaternary North American Vegetational History 305

Comparison Chart South-Central and Interior Alaska

Figure 8.25. Comparison chart of Late Quaternary pollen zones in south-central and interior Alaska. Dashed lines rep-
resent uncertainty about boundary positions. Reprinted from Ager and Brubaker (1985, fig. 5) with the permission of
the American Association of Stratigraphic Palynologists Foundation.

ence a similar pattern was likely followed during each of
the 18-20 glacial-interglacial transitions of the past 1.6
m.y. The sequence was a herb zone bordering the retreating
glacial margin (tundra to park tundra); spruce, fir, and
other boreal species; pine (climatic optimum); and dedicu-
ous forest (oak with hemlock, birch, and beech in the
northern region).
The boreal coniferous forest had its primary center of
origin in the highlands of the northern Rocky Mountains in
the Middle Eocene (Abies, Chamaecyparis, Picea, Pinus,
Pseudolarix, Thuja, Tsuga, Betula). Beginning at -3.4 Ma it
began expanding extensively into the lowlands and subse-
quently alternated between numerous regional eradica-
tions by glacial advances and reintroductions during inter-
glacial times. The current version dates from the retreat of
the last glaciers at ~6 Kya, and some areas have supported
boreal coniferous forest for only 3 Ky.
In the Plains area, trees diminished to the point where
the vegetation could be designated a grassland near the
onset of extensive Arctic glaciation at -2.4 Ma. After that
time the southern region broadly supported oak-hickory
savanna alternating with increased deciduous species
mixed with some spruce. Southern coastal pines also ex-
tended inland with cooler climates. Along the western
margin pine-juniper and Artemisia periodically extended
onto the grassland. In the north, the ecotone was with the
boreal forest (especially Picea); along the eastern margin it
was with the deciduous forest.
In one respect vegetational history in the west and south-
western United States parallels the observations of Davis
(1976) for the deciduous forest of the east: the present-day
widespread warm deserts (Chihuahuan, Sonoran, Mojave)
and desert grasslands are the exceptions for the past 1.6
m.y. For most of this time the region was characterized by
low-pressure systems, higher rainfall, moderate tempera-
tures, numerous lakes, lower forest ecotones, and savanna -
woodlands (e.g., juniper-pinon pine; evergreen oak-lau-
rel-madrone; Artemisia steppe) in the presently desert
lowlands. Desert elements, in turn, were mostly restricted
to edaphically or topographically favorable habitats. Cali-
fornia chaparral and coastal sage communities developed
from Pliocene predecessors that grew as sclerophyllous
shrubs in the understory of oak-laurel-madrone wood-
lands (Axelrod, 1978, 1989). Some of these were derived
from plants of the Sonoran Desert of northern Mexico that
grew in marginal areas of greater summer rainfall. Others
306 Late Cretaceous and Cenozoic History of North American Vegetation
came from temperate forests to the north: Aesculus
palmeri, Baccharis sergiloides, Fraxinus trifoliata, Mal-
osma laurina, Rhus integrifolia, Sanicula deserticola, and
Satureja chandler!. Their coalescence into a distinct recog-
nizable community was favored by the Mediterranean
(summer dry) climates and fires that increased with the
first Aftonian interglacial period of aridity and by subse-
quent hypsithermal intervals, including the last one be-
tween 8 and 4 Kya. Some communities had different com-
binations of ecologically compatible taxa after each
reshuffling, and this fluidity has probably characterized
western North American vegetation at the association level
throughout its history.
At about 3.4 Ma the Arctic tundra included an esti-
mated 1500 species from alpine and cool-temperate prede-
cessors that had gradually become adapted to the light,
temperature, and precipitation regimes of the far north.
With the rigors of subsequent climatic changes the number
has been reduced to less than 1100 species, of which about
700 occur in the North American Arctic (Love and Love,
1974).
Little is known about the Early Quaternary history of
tropical elements in peninsular Florida. It is assumed that
Rhizophora and associated taxa increased in each of the
warm intervals and declined or disappeared during cold
times.
The temperature difference in MAT in much of North
America between the last glacial maximum at 18 Kya and
the climatic optimum at ~7 Kya was ~10°C. Recall from the
paleotemperature curve (Fig. 3.1) that this is approximately
the average change in benthic temperatures during the 70-
m.y. span between the Late Cretaceous and the beginning
of the Pleistocene. A direct comparison is misleading, but
it is clear that the pace and intensity of climatic change in-
creased in Quaternary times. Three of the most extensive
North American plant formations (desert, grassland, tun-
dra) appear in this period and are among the most recent to
develop in modern form. The current versions date back
mostly to the end of the last glaciation. At ~6 Kya (the mid-
point of the current 10-12 Kya interglacial) they began re-
sponding to a new cooling phase, but this trend has been
countered in the past 200 years by global warming.
The Quaternary is the interval in which paleoecology
interfaces most closely with the environments and pro-
cesses operating in the modern biota. Thus, our recon-
struction of Quaternary vegetational history is more com-
plete than for any previous time. In turn, this history
provides a better understanding of the modern flora and
constitutes the only basis available for estimating its fu-
ture. Although the information is comparatively detailed,
the complexities of environmental change and biotic re-
sponse still exceed our capacity to describe them precisely.
In particular, even though some components of paleovege-
tation analysis have been improved and can be applied
with great precision, others allow considerable latitude in
interpretation. For example, the ability to identify plant mi-
crofossils mostly to genus or family is especially limiting
when considering vegetation types often characterized by
different species of Pinus, Cyperaceae, Gramineae, Acer,
Compositae, Betula, Ericaceae, Quercus, and others. In this
connection megafossils are valuable and context informa-
tion is essential.
Even with these limitations, however, the last decade
has produced results that are both revolutionary and fun-
damental to our understanding of biotic history:
1. climates switch modes much faster than previously
realized;
2. changes in climate first evident in ice cores and ma-
rine sediments from the North Atlantic may be paral-
leled in the North Pacific and hemispheric in scope;
3. the present warm interglacial is by far the exceptional
climate during the past 1.6 m.y.;
4. vegetation characteristic of extensive regions at pres-
ent (deciduous forest in the east, desert in the south-
west) is not typical of the past 1.6 m.y. when montane
coniferous forest and juniper-pinon pine woodland
and steppe occupied much of these areas; and
5. organisms respond to environmental change as indi-
viduals.
Many of these generalizations are discernable in the Ter-
tiary record, but they are more evident in the Quaternary
where the pace was faster and the record is clearer. The av-
erage migratory rate for Picea in the Holocene of eastern
North America was 14.1 km/century; Quercus 12.6; north-
ern species of Pinus 13.5; and southern pines 8.1 (Delcourt
and Delcourt, 1987). Where genetically controlled physio-
logical processes are similar among taxa, the organisms po-
tentially may have similar past histories; but even this po-
tential is subject to the vicissitudes of barriers and seed
dispersal, edaphic factors, anthropogenic influence, and
epidemic disease.
Realization of the temporal nature of communities has
implications for such time-honored concepts as succes-
sion, climax, and geofloras. If climates change on a scale
more rapidly than implied by the older four-part glacial
chronology, then designating the vegetation type character-
istic of a given region loses much of its meaning, especially
for the turbulent times of glacial-interglacial transitions. It
also reduces the value of a concept that envisions the
movements of vegetation as large intact blocks. On the
other hand, it is worth noting that the number of taxa hav-
ing similar ecological and dispersal potentials increases
with higher units of vegetation. Thus, at the association
level there are communities that are rare or absent in the
modern biota (spruce-ash parkland in the upper midwest,
sagebrush-pine woodland in the southwest). At the same
time there are groupings at the formation level, which are
frequently recognized by Tertiary paleontologists, that
have persisted for long periods of time. Discussion of such
conceptual matters improves with better definition of the
time span and the hierarchical level being considered.
The actual association of elements in a community, es-

pecially when they represent novel combinations based on
a mixture of wind-blown pollen types, usually can only be
assummed. The diagrams presented are a general reflection
of major trends in the history of a limited number of com-
ponents, and interpretations can easily go beyond the sen-
sitivity of the method. One approach to Late Neogene and
Quaternary vegetational history is to begin with the as-
sumption that the modern vegetation has persisted rela-
tively unchanged back through time. A change is accepted
only when the data actually force that conclusion. This is
in contrast to a mind-set that anticipates the dramatic and
leads to quick acceptance of evidence in apparent support
of a dynamicism that approaches chaos and randomness,
be it sizable amounts of spruce and fir pollen in Texas bogs
or Arctic tundra and glacial striations on rocks in the
southern Appalachian Mountains. As noted previously, the
first makes for sound science but it can delay recognition
and acceptance of the unexpected. The second makes for
good theater, but it is more receptive to new ideas and in-
novative interpretations. Even though a conservative route
has been followed in this summary, the vegetation of North
America still emerges as dynamic and immensely com-
plex. Its history is intriguing to study and offers a great deal
to contemplate about future biotic changes.
Notes
1. Comparable glacial stages in the European Alps are
named for four tributaries of the Danube west of Munich:
the Gtinz, Mindel, Riss, and Wiirm. The interglacials are
conventionally designated as G/M (Giinz/Mindel), M/R,
and R/W or Eemian (the Sangamon interglacial in North
America).
2. On the surface of several of the Texas bogs there is an
alga (Zygogonium ericetorum) belonging to the Chloro-
phyta that is uniquely purple. Ralph Alston, coauthor with
Billie Turner of the classic text on Biochemical Systemat-
ics, was interested in these pigments and in 1957 he ac-
companied me on a collecting trip to the Soefje bog near
Palmetto State Park. While I was coring and he was collect-
ing algae, a bull approached, making all the signs of de-
fending his territory. I told Dr. Alston I thought we should
leave, but he said it was a bluff. The bull charged and we
barely made it out by throwing corer, rods, vials, and plant
presses over a barbed wire fence and diving over ourselves.
Later I received a reprint with the inscription: "When you
see this, you will remember that in the tradition of great
scientists the world over / was willing to risk disfigure-
ment, even death, to collect this material while you cringed
behind the protection of a barbed wire fence clutching a
rare orchid which, as I suspected all along, turned out to be
an Ms." Ralph Alston died early in his career shortly after-
ward, and the reprint is a treasured memento of a re-
spected scientist and generous mentor.
3. The question of "significant" versus "moderate"
changes in the vegetation of southeastern United States has
been long debated because the terms are relative and de-
pend, in part, on perspective. Specialists may see events as
profound and dramatic, while generalists see them in a
broader context. This opens the opportunity for endless
Quaternary North American Vegetational History 307
discussions and, in hindsight, amusing debates. As a re-
cent Master's student from the University of Texas, I made
my first foray into the national meetings scene at the Amer-
ican Institute of Biological Sciences conference at Indiana
University in 1958. My paper was on a comparison of
Potzger and Tharp's results and my own findings from the
Texas bogs. When the session was opened for questions
someone, whom I later found to be Dan Livingstone, pro-
vided a baptism of fire with a critique that covered method-
ology, results, interpretation, and conclusions. Another
member of the audience, Paul Martin, waded in with a de-
fense. As I watched these titans shred and reassemble my
paper, I recall concluding that research in paleoecology
was going to be anything but dull. Afterward, Livingstone
said, "Nice paper."
In 1964 I was asked by the Society for the Study of Evo-
lution to present a summary of the vegetational history of
the southeastern United States at its meetings in Chapel
Hill. Coming from the perspective of a specialist in the Ter-
tiary, where the vegetation included extinct genera and el-
ements from tropical America, the Old World tropics, and
eastern Asia, I noted that, within this context, the changes
in the biota of the southeastern United States were not ex-
tensive (Graham, 1965). Those specializing in the Quater-
nary correctly regarded this as an understatement. Follow-
ing the Chapel Hill meeting I was invited to present the
paper at a midwestern university where two colleagues
had orchestrated a response. They concluded their critique
with the revelation that rocks with purportedly glacial stri-
ations had just been discovered in the southern Ap-
palachian Mountains. However, these striations were un-
usual in being all about the same diameter, U shaped in
outline, and coated with iron oxide. They turned out to be
cable grooves made during old lumbering operations as
logs were dragged over the granite rocks. Periglacially
modified strata do occur locally at high elevations in the
northern and central Appalachian Mountains (Clark, 1968;
Pewe, 1983).
References
Adam, D. P. 1988. Palynology of two Upper Quaternary
cores from Clear Lake, Lake County, California. U.S.
Geological Survey, Reston, VA. Professional Paper
1363, pp. 1-86.
Agassiz, L. 1840. Etudes sur les Glaciers. Privately pub-
lished essay, Neuchatel, Switzerland.
Ager, T. A. and L. B. Brubaker. 1985. Quaternary palynol-
ogy and vegetational history of Alaska. In: V. M.
Bryant, Jr. and R. G. Holloway (eds.), Pollen Records of
Late-Quaternary North American Sediments. Ameri-
can Association of Stratigraphic Palynologists Founda-
tion, Dallas, TX. pp. 353-384.
Aguirre, E. and G. Pasini. 1985. The Pliocene-Pleistocene
boundary. Episodes 8: 116-120.
Allison, T. D., R. E. Moeller, and M. B. Davis. 1986. Pollen
in laminated sediments provides evidence for a mid-
Holocene forest pathogen outbreak. Ecology 67: 1101-
1105.
Ammann, B. R. 1977. A pollen morphological distinction
between Finns banksiana lamb, and P. resinosa ait.
Pollen Spores 19: 521-529.
Anderson, R. S. 1990. Holocene forest development and
308 Late Cretaceous and Cenozoic History of North American Vegetation
paleoclimates within the central Sierra Nevada, Cali-
fornia. J. Ecol. 78: 470-489.
and T. R. Van Devender. 1991. Comparison of pollen
and macrofossils in packrat (Neotoma) middens: a
chronological sequence from the Waterman Mountains
of southern Arizona, U.S.A. Rev. Palaeobot. Palynol.
68: 1-28.
Anderson, T. W. 1985. Late-Quaternary pollen records from
eastern Ontario, Quebec, and Atlantic Canada. In: V.
M. Bryant, Jr. and R. G. Holloway (eds.), Pollen
Records of Late-Quaternary North American Sedi-
ments. American Association of Stratigraphic Palynol-
ogists Foundation, Dallas, TX. pp. 281-326.
Antevs, E. 1938. Postpluvial climatic variation in the
southwest. Am. Meteorl. Soc. Bull. 19:190-193.
Atwater, B. E et al. (+ 8 authors). 1986. A fan dam for Tulare
Lake, California, and implications for the Wisconsin
glacial history of the Sierra Nevada. Geol. Soc. Am.
Bull. 97: 97-109.
Axelrod, D. I. 1978. The origin of coastal sage vegetation,
Alta and Baja California. Am. J. Bot. 65: 1117-1131.
. 1989. Age and origin of chaparral. Nat. Hist. Mus.
LA County, Sci. Ser. 34: 7-19.
Baker, R. G. and K. A. Wain. 1985. Quaternary pollen
records from the Great Plains and central United
States. In: V. M. Bryant, Jr. and R. G. Holloway (eds.),
Pollen Records of Late-Quaternary North American
Sediments. American Association of Stratigraphic Pa-
lynologists Foundation, Dallas, TX. pp. 191-203.
Baker, R. G. et al. (+ 6 authors). 1986. A full-glacial biota
from southeastern Iowa. J. Quatern. Sci. 1: 91-107.
Barnosky, C. W. 1985. Late Quaternary vegetation in the
southwestern Columbia Basin, Washington. Quatern.
Res. 23: 109-122.
, P M. Anderson, and R J. Bartlein. 1987a. The north-
western U.S. during deglaciation; vegetational history
and paleoclimatic implications. In: W. E Ruddiman
and H. E. Wright, Jr. (eds.), North America and Adja-
cent Oceans during the Last Deglaciation. The Geology
of North America, Decade of North American Geology.
Geological Society of America, Boulder, CO. Volume
K-3,pp. 289-321.
Barnowsky, C. W, E. C. Grimm, and H. E. Wright, Jr. 1987b.
Towards a postglacial history of the northern Great
Plains: a review of paleoecologic problems. Ann. Car-
negie Mus. 56: 259-273.
Barrie, J. V, K. W. Conway, R. W. Mathewes, H. W. Josen-
hans, and M. J. Johns. 1994. Submerged Late Quater-
nary terrestrial deposits and paleoenvironment of
northern Hecate Strait, British Columbia continental
shelf, Canada. Quatern. Int. 20: 123-129.
Barry, R. G. 1983. Late-Pleistocene climatology. In: S. C.
Porter (ed.), Late-Quaternary Environments of the
United States. The Late Pleistocene. University of
Minnesota Press, Minneapolis, MN. Vol. 1, pp. 390-
407.
Bartlein, R J. 1988. Late-Tertiary and Quaternary paleoenvi-
ronments. In: B. Huntley and T. Webb III (eds.), Vegeta-
tion History. Kluwer Academic Publishers, Dordrecht,
pp. 113-152.
, T. Webb III, and E. Fieri. 1984. Holocene climatic
change in the northern midwest: pollen-derived esti-
mates. Quatern. Res. 22: 361-374.
Bell, M. and E. R Laine. 1985. Erosion of the Laurentide re-
gion of North America by glacial and glaciofluvial
processes. Quatern. Res. 23: 154-174.
Benedict, J. B. 1973. Chronology of cirque glaciation, Col-
orado Front Range. Quatern. Res. 3: 584-599.
Benninghoff, W. S. 1962. Calculation of pollen and spore
density in sediments by addition of exotic pollen in
known quantities. Pollen Spores 4: 332-333.
Benson, L. J. Burdett, S. Lund, M. Kashgarian, and S.
Mensing. 1997. Nearly synchronous climate change in
the Northern Hemisphere during the last glacial termi-
nation. Nature 388: 263-265.
Benson, L. and R. S. Thompson. 1987. The physical record
of lakes in the Great Basin. In: W. E Ruddiman and
H. E. Wright, Jr. (eds.), North America and Adjacent
Oceans during the Last Deglaciation. The Geology of
North America, Decade of North American Geology,
Geological Society of America, Boulder, CO. Vol. K-3,
pp. 241-260.
Betancourt, J. L. and T. R. Van Devender. 1981. Holocene
vegetation in Chaco Canyon, New Mexico. Science
214:656-658.
Birks, H. J. B. 1976. Late-Wisconsinan vegetational history
at Wolf Creek, central Minnesota. Ecol. Monogr. 46:
395-429.
. 1993. Quaternary palaeoecology and vegetation
science—current contributions and possible future
developments. Rev. Palaeobot. Palynol. 79: 153-177.
and A. D. Gordon. 1985. Numerical Methods in
Quaternary Pollen Analysis. Academic Press, New
York.
Blaise, B., J. J. Clague, and R. W. Mathewes. 1990. Time of
maximum Late Wisconsin glaciation, west coast of
Canada. Quatern. Res. 34: 282-295.
Bond, G. et al. (+ 13 authors). 1992. Evidence for massive
discharges of icebergs into the North Atlantic Ocean
during the last glacial period. Nature 360: 245-249.
Bowen, D. Q. 1978. Quaternary Geology, A Stratigraphic
Framework for Multidisciplinary Work. Pergamon
Press, Oxford, U.K. [See also later reprints and addi-
tions to 1988.]
Bradbury, J. P. 1997a. A Diatom record of climate and hy-
drology for the past 200ka from Owens Lake, Califor-
nia with comparison to other Great Basin records.
Quaternary Sci. Rev. 16: 203-219.
. 1997b. A diatom-based paleohydrologic record of
climate change for the past 800 k.y. from Owens Lake,
California. Geol. Soc. Am. Sp. Paper 317: 99-112.
Bradbury, J. P. and W. E. Dean (eds.). 1993. Elk Lake, Min-
nesota: Evidence for Rapid Climate Change in the
North-Central United States. Geological Society of
America, Boulder, CO. Special Paper 276.
Broccoli, A. J. and S. Manabe. 1987. The effects of the Lau-
rentide ice sheet on North American climate during
the last glacial maximum. Geograph. Phys. Quatern.
41: 291-299.
Broecker, W. S. and J. van Donk. 1970. Insolation changes,
ice volumes, and the O18 record in deep-sea cores. Rev.
Geophys. Space Phys. 8:169-197.
Brown, C. A. 1938. The flora of Pleistocene deposits in the
western Florida Parishes, west Feliciana Parish, and
east Baton Rouge Parish, Louisiana. In: H. N. Fisk,
H. G. Richards, C. A. Brown, and W. C. Steere (eds.).
Contributions to the Pleistocene History of the Florida
Parishes of Louisiana. Department of Conservation,

Louisiana Geological Survey, New Orleans, LA. Geo-
logical Bull. 12, pp. 59-96.
Brunner, C. A. 1982. Paleoceanography of surface waters in
the Gulf of Mexico during the Late Quaternary.
Quatern. Res. 17: 105-119.
Bryant, V. M., Jr., 1977. A 16,000 year pollen record of veg-
etational change in central Texas. Palynology 1:
143-156.
and R. G. Holloway. 1985. A Late Quaternary paleo-
environmental record of Texas: an overview of the
pollen evidence. In: V. M. Bryant, Jr. and R. G. Hol-
loway (eds.), Pollen Records of Late Quaternary North
American Sediments. American Association of Strati-
graphic Palynologists Foundation, Dallas, TX. pp.
39-70.
Bryant, V. M., Jr. and H. J. Shafer. 1977. The Late Quater-
nary paleoenvironment of Texas: a model for the
archeologist. Bull. Texas Archeol. Soc. 48:1-25.
Carrara, P E. 1989. Late Quaternary glacial and vegetative
history of the Glacier National Park region, Montana.
U.S. Geol. Surv. Bull. 1902:1-64.
, W. N. Mode, M. Rubin, and S. W. Robinson. 1984.
Deglaciation and postglacial timberline in the San
Juan Mountains, Colorado. Quatern. Res. 21: 42-55.
Chapman, J., P. A. Delcourt, P. A. Cridlebaugh, A. B. Shea,
and H. R. Delcourt. 1982. Man-land interaction:
10,000 years of American Indian impact on native
ecosystems in the Lower Little Tennessee River Valley,
eastern Tennessee. Southeast. Archaeol. 1:115-121.
Clark, G. M. 1968. Sorted patterned ground: new Ap-
palachian localities south of the glacial border. Sci-
ence 161: 355-356.
Clark, P. U. 1995. Fast glacier flow over soft beds. Science
267:43-45.
CLIMAP Project Members. 1976. The surface of the ice-age
earth. Science 191: 1131-1137.
Clisby, K. H. and P. B. Sears. 1956. San Agustin Plains —
Pleistocene climatic changes. Science 124: 537-538.
COHMAP Members. 1988. Climatic changes of the last
18,000 years: observations and model simulations.
Science 241: 1043-1052.
Cole, K. 1983. Late Pleistocene vegetation of Kings Canyon,
Sierra Nevada, California. Quatern. Res. 19: 117-129.
. 1985. Past rates of change, species richness, and a
model of vegetational inertia in the Grand Canyon,
Arizona. Am. Nat. 125: 289-303.
Colinvaux, P. A. 1963. A pollen record from Arctic Alaska
reaching glacial and Bering land bridge times. Nature
198: 609-610.
. 1964a. Origin of ice ages: pollen evidence from
Arctic Alaska. Science 145: 707-708.
. 1964b. The environment of the Bering land bridge.
Ecol. Monogr. 34: 297-329.
1996. Low-down on a land bridge. Nature 382:
21-23.
Coplen, T. B., I. J. Winograd, J. M. Landwehr, and A. C.
Riggs. 1994. 500,000-year stable carbon isotopic
record from Devils Hole, Nevada. Science 263: 361-
365.
Corridor essays. 1996. Quatern. Int. 32.
Cridlebaugh, R A. 1984. American Indian and Euro-Amer-
ican impact on Holocene vegetation in the Lower Lit-
tle Tennessee River Valley, east Tennessee. Ph.D. dis-
sertation, University of Tennessee, Knoxville, TN.
Quaternary North American Vegetational History 309
Critchfield, W. B. and E. L. Little, Jr. 1966. Geographic dis-
tribution of the pines of the world. U.S. Forest Service,
Washington, D.C. Miscellaneous Publ. 991.
Cuffey, K. M. et al. (+ 5 authors). 1995. Large Arctic tem-
perature change at the Wisconsin—Holocene transi-
tion. Science 270: 455-458.
Davis, J. H., Jr. 1946. The peat deposits of Florida, their oc-
currence, development, and uses. Fla. Dept. Conserv.,
Geol. Surv., Geol. Bull. 30:1-247.
Davis, M. B. 1963. On the theory of pollen analysis. Am. J.
Sci. 261: 897-912.
. 1965. A method for determination of absolute
pollen frequency. In: B. Kummel and D. Raup (eds.),
Handbook of Paleontological Techniques. W. H. Free-
man and Co., San Francisco, CA. pp. 674-686.
. 1967. Pollen accumulation rates at Rogers Lake,
Connecticut, during late- and postglacial time. Rev.
Palaeobot. Palynol. 2: 219-230.
. 1969. Climatic changes in southern Connecticut
recorded by pollen deposition at Rogers Lake. Ecology
50: 409-422.
. 1976. Pleistocene biogeography of temperate de-
ciduous forests. Geosci. Man 13: 13-26.
. 1981a. Outbreaks of forest pathogens in Quaternary
history. In: Proceedings of the IVth International Paly-
nologists Conference, Lucknow, 1976-1977, Vol. 3, pp.
216-227.
1981b. Quaternary history and the stability of forest
communities. In: D. C. West, H. H. Shugart, and D. B.
Botkin (eds.), Forest Succession: Concepts and Appli-
cations. Springer-Verlag, Berlin, pp. 132-153.
. 1983. Holocene vegetational history of eastern
United States. In: H. E. Wright, Jr. (ed.), Late-Quaternary
Environments of the United States, The Holocene.
University of Minnesota Press, Minneapolis, MN. Vol.
2, pp. 166-181.
. 1989. Insights from paleoecology on global change.
Ecol. Soc. Am. Bull. 70: 222-228.
-, L. B. Brubaker, and T. Webb III. 1973. Calibration of
absolute pollen influx. In: H. J. B. Birks and R. G. West
(eds.), Quaternary Plant Ecology. Blackwell Science
Publishers, Oxford, U.K. pp. 9-25.
Davis, M B., K. D. Woods, S. L. Webb, and R. P. Futyma.
1986. Dispersal versus climate: expansion of Fagus
and Tsuga into the upper Great Lakes region. Vegetatio
67: 93-103.
Dawson, A. G. 1992. Ice Age Earth, Late Quaternary Geol-
ogy and Climate. Routledge, London.
Deevey, E. S., Jr. 1939. Studies on Connecticut Lake Sedi-
ments. I. A postglacial climatic chronology for south-
ern New England. Am. J. Sci. 237: 691-724.
. 1949. Biogeography of the Pleistocene. Bull. Geol.
Soc. Amer. 60: 1315-1416.
Delcourt, H. R. 1979. Late Quaternary vegetation history of
the eastern Highland Rim and adjacent Cumberland
Plateau of Tennessee. Ecol. Monogr. 49: 255-280.
. 1987. The impact of prehistoric agriculture and
land occupation on natural vegetation. Trends Ecol.
Evol. 2: 39-44.
and P. A. Delcourt. 1975. The Blufflands: Pleis-
tocene pathway into the Tunica Hills. Am. Midland
Nat. 94: 385-400.
. 1985a. Comparison of taxon calibrations, modern
analogue techniques, and forest-stand simulation mod-
310 Late Cretaceous and Cenozoic History of North American Vegetation
els for the quantitative reconstruction of past vegeta-
tion. Earth Surf. Processes Landforms 10: 293-304.
—. 1985b. Quaternary palynology and vegetational
history of the southeastern United States. In: V. M.
Bryant, Jr. and R. G. Hollo way (eds.), Pollen Records of
Late-Quaternary North American Sediments. Ameri-
can Association of Stratigraphic Palynologists Founda-
tion, Dallas, TX. pp. 1-37.
—. 1986. Late Quaternary vegetational change in the
central Atlantic States. In: J. N. McDonald and S. O.
Bird (eds.), The Quaternary of Virginia. Virginia Divi-
sion of Mineral Resources, Charlottesville, VA. Publi-
cation 75, pp. 23-35.
-. 1991. Late-Quaternary vegetation history of the In-
terior Highlands of Missouri, Arkansas, and Okla-
homa. In: D. Henderson and L. D. Hedrick (eds.),
Restoration of Old Growth Forests in the Interior High-
lands of Arkansas and Oklahoma. Ouachita National
Forest and Winrock International Institute for Agricul-
tural Development, Morrilton, AR. pp. 15-30.
Delcourt, P. A. 1980. Goshen Springs: Late Quaternary veg-
etation record for southern Alabama. Ecology 61:
371-386.
and H. R. Delcourt. 1977. The Tunica Hills,
Louisiana—Mississippi: late glacial locality for spruce
and deciduous forest species. Quatern. Res. 7: 218-237.
. 1984. Late Quaternary paleoclimates and biotic re-
sponses in eastern North America and the western
North Atlantic Ocean. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 48: 263-284.
. 1987. Long-Term Forest Dynamics of the Temperate
Zone: A Case Study of Late-Quaternary Forests in East-
ern North America, Ecological Studies 63. Springer-
Verlag, Berlin.
. 1993. Paleoclimates, paleovegetation, and paleoflo-
ras during the Late Quaternary. In: Flora of North
America Editorial Committee (eds.), Flora of North
America North of Mexico. Oxford University Press,
Oxford, U.K. Vol. 1, pp. 71-94.
. In press. Quaternary paleoecology of the lower
Mississippi Valley. Eng. Geol.
, R. C. Brister, and L. E. Lackey. 1980. Quaternary
vegetation history of the Mississippi Embayment.
Quatern. Res. 13: 111-132.
, P. A. Cridlebaugh, and J. Chapman. 1986. Holocene
ethnobotanical and paleoecological record of human
impact on vegetation in the Little Tennessee River Val-
ley, Tennessee. Quatern. Res. 25: 330-349.
-, and T. Webb III. 1984. Atlas of Mapped Distribu-
tions of Dominance and Modern Pollen Percentages
for Important Tree Taxa of Eastern North America.
American Association of Stratigraphic Palynologists
Foundation, Dallas, TX. Contrib. Ser. 14.
Dillon, L. S. 1956. Wisconsin climate and life zones in
North America. Science 123: 167-176.
Dimbleby, G. W. 1985. The Palynology of Archaeological
Sites. Academic Press, New York.
Edwards, M. E. and E. D. Barker, Jr. 1994. Climate and veg-
etation in northeastern Alaska. Palaeogeogr., Palaeocli-
matol., Palaeoecol. 109:127-135.
Edwards, M. E., J. C. Dawe, and W. S. Armbruster. 1991.
Pollen size of Betula in northern Alaska and the inter-
pretation of Late Quaternary vegetation records. Can. J.
Bot. 69: 1666-1672.
Edwards, R. L. 1995. Technical Comments—paleotopogra-
phy of glacial-age ice sheets. Science 267: 536 [see also
Response, Peltier, 1995].
Elias, S. A., S. K. Short, C. H. Nelson, and H. H. Birks. 1996.
Life and times of the Bering land bridge. Nature 382:
60-63.
Elias, T. S. 1980. The Complete Trees of North America,
Field Guide and Natural History. Van Nostrand Rein-
hold, New York.
Elsik, W. C. 1969. Late Neogene palynomorph diagrams,
northern Gulf of Mexico. Trans. Gulf Coast Assoc.
Geol. Soc. 19: 509-528.
Emiliani, C. 1955. Pleistocene temperatures. J. Geol. 63:
538-578.
Engstrom, D. R., B. C. S. Hansen, and H. E. Wright, Jr. 1990.
A possible Younger Dryas record in southeastern
Alaska. Science 250: 1383-1385.
Ewing, M. and W. L. Donn. 1956. A theory of ice ages. Sci-
ence 123: 1061-1066.
Faegri, K., R Kaland, and K. Krzywinski. 1989. Textbook of
Pollen Analysis. Wiley, New York. 4th edition.
Fairbanks, R. G. 1989. A 17,000-year glacio-eustatic sea
level record: influence of glacial melting rates on the
Younger Dryas event and deep-ocean circulation. Na-
ture 342: 637-642.
Feng, X. and S. Epstein. 1994. Climatic implications of an
8000-year hydrogen isotope time series from bristle-
cone pine trees. Science 265:1079-1081.
Flint, R. F and M. Rubin. 1955. Radiocarbon dates of pre-
Mankato events in eastern and central North America.
Science 121: 649-658.
Gajewski, K. 1993. The role of paleoecology in the study of
global climate change. Rev. Palaeobot. Palynol. 79:
141-151.
, S. Payette, and J. C. Ritchie. 1993. Holocene vegeta-
tion history at the boreal-forest-shrub-tundra transi-
tion in north-western Quebec. J. Ecol. 81: 433-443.
Gallimore, R. G. and J. E. Kutzbach. 1996. Role of orbitally
induced changes in tundra area in the onset of glacia-
tion. Nature 381: 503-505.
Gaudreau, D. C. and T. Webb III. 1985. Late-Quaternary
pollen stratigraphy and isochrone maps for the north-
eastern United States. In: V. M. Bryant, Jr. and R. G.
Holloway (eds.), Pollen Records of Late-Quaternary
North American Sediments. American Association of
Stratigraphic Palynologists Foundation, Dallas, TX.
pp. 247-280.
Graham, A. 1965. Origin and evolution of the biota of
southeastern North America: evidence from the fossil
plant record. Evolution 18: 571-585.
and C. Heimsch. 1960. Pollen studies of some Texas
peat deposits. Ecology 41: 785-790.
Graham, R. W. and J. I. Mead. 1987. Environmental fluctu-
ations and evolution of mammalian faunas during the
last deglaciation in North America. In: W. E Ruddiman
and H. E. Wright, Jr. (eds.), North America and Adja-
cent Oceans during the Last Deglaciation. The Geology
of North America, Decade of North American Geology.
Geological Society of America, Boulder, CO. Vol. K-3,
pp. 371-402.
Gray, J. (Coordinator). 1965. Techniques in palynology. In:
B. Kummel and D. Raup (eds.), Handbook of Paleonto-
logical Techniques. W. H. Freeman and Co., San Fran-
cisco, CA. pp. 471-706.

and W. Smith. 1962. Fossil pollen and archaeology.
Archaeology 15: 16-26.
Grimm, E. C. 1983. Chronology and dynamics of vegetation
change in the prairie-woodland region of southern
Minnesota, U.S.A. New Phytol. 93: 311-350.
. 1988. Data analysis and display. In: B. Huntley and
T. Webb III (eds.), Vegetation History. Kluwer Aca-
demic Publishers, Dordrecht, pp. 43-76.
-, G. L. Jacobson, Jr., W. A. Watts, B. C. S. Hansen,
and K. A. Maasch. 1993. A 50,000-year record of cli-
mate oscillations from Florida and its temporal corre-
lation with the Heinrich events. Science 261: 198-
200.
Groot, J. J. 1991. Palynological evidence for Late Miocene,
Pliocene and Early Pleistocene climate changes in the
middle U.S. Atlantic Coastal Plain. Quatern. Sci. Rev.
10: 147-162.
Griiger, J. 1973. Studies on the Late Quaternary vegetation
history of northeastern Kansas. Geol. Soc. Am. Bull.
84: 239-250.
Guilderson, T. E, R. G. Fairbanks, and J. L. Rubenstone. 1994.
Tropical temperature variations since 20,000 years ago:
modulating interhemispheric climatic change. Science
263: 663-665.
Hafsten, U. 1964. A standard pollen diagram for the south-
ern High Plains, USA, covering the period back to the
Early Wisconsin glaciation. In: Report of the Vlth In-
ternational Congress on the Quaternary, Warsaw, 1961,
Vol. II, Palaeobotanical Sect., pp. 407-420.
Hall, S. A. 1977. Late Quaternary sedimentation and paleo-
ecologic history of Chaco Canyon, New Mexico. Geol.
Soc. Am. Bull. 88: 1593-1618.
. 1981. Holocene vegetation at Chaco Canyon: pollen
evidence from alluvium and pack rat middens. In: So-
ciety of American Archeology. 46th Annual Meeting,
San Diego, 1981. Program and Abstracts, p. 61.
. 1982. Reconstruction of local and regional
Holocene vegetation in the arid southwestern United
States based on combined pollen analytical results
from Neotoma middens and alluvium. In: Interna-
tional Union for Quaternary Research, XI Congress,
Moscow, Vol. 1, p. 130.
. 1985. Quaternary pollen analysis and vegetational
history of the southwest. In: V. M. Bryant, Jr. and R. G.
Holloway (eds.), Pollen Records of Late-Quaternary
North American Sediments. American Association of
Stratigraphic Palynologists Foundation, Dallas, TX.
pp. 95-123.
. 1990. Holocene landscapes of the San Juan Basin,
New Mexico; geomorphic, climatic, and cultural dy-
namics. In: N. E Lasca and J. Donahue (eds.), Archaeo-
logical Geology of North America. Geological Society
of America, Boulder, CO. Centennial Special Vol. 4.
pp. 323-334.
and S. Valastro, Jr. 1995. Grassland vegetation in the
southern Great Plains during the last glacial maxi-
mum. Quatern. Res. 44: 237-245.
Hallberg, G. R. 1986. Pre-Wisconsin glacial stratigraphy of
the central plains region in Iowa, Nebraska, Kansas,
and Missouri. In: V. Sibrava, D. Q. Bowen, and G. M.
Richmond (eds.), Quaternary Glaciations in the North-
ern Hemisphere. Report of the International Geologi-
cal Correlation Programme Project 24. Pergamon Press,
Oxford, U.K. pp. 11-15.
Quaternary North American Vegetational History 311
Hansen, H. E 1939. Pollen analysis of a bog near Spokane,
Washington. Bull. Torrey Bot. Club. 66: 215-220.
. 1947. Postglacial forest succession, climate and
chronology in the Pacific Northwest. Am. Philos. Soc.
37: 1-130.
Hays, J. D., J. Imbrie, and N. J. Shackleton. 1976. Variations
in the earth's orbit: pacemaker of the ice ages. Science
194:1121-1132.
Hebda, R. J. 1995. British Columbia vegetation and climate
history with focus on 6 Kya BE Geograph. Phys.
Quatern. 49: 55-79.
Hedges, J. I., J. R. Ertel, and E. B. Leopold. 1982. Lignin geo-
chemistry of a Late Quaternary sediment core from
Lake Washington. Geochim. Cosmochim. Acta 46:
1869-1877.
Heusser, C. J. 1952. Pollen profiles from southeastern
Alaska. Ecol. Monogr. 22: 331-352.
. 1977. Quaternary palynology of the Pacific slope of
Washington. Quatern. Res. 8: 282-306.
. 1985. Quaternary pollen records from the interior
Pacific Northwest coast: Aleutians to the Oregon -
California boundary. In: V. M. Bryant, Jr. and R. G. Hol-
loway (eds.), Pollen Records of Late-Quaternary North
American Sediments. American Association of Strati-
graphic Palynologists Foundation, Dallas, TX. pp. 141-
165.
-. 1990. Late Quaternary vegetation of the Aleutian
Islands, southwestern Alaska. Can. J. Bot. 68: 1320-
1326.
Heusser, L. E. 1978. Pollen in Santa Barbara Basin, Califor-
nia: a 12,000-yr record. Geol. Soc. Am. Bull. 89:
673-678.
and J. E. King. 1988. North America, with special
emphasis on the development of the Pacific coastal
forest and prairie/forest boundary prior to the last
glacial maximum. In: B. Huntley and T. Webb III (eds.),
Vegetation History. Kluwer Academic Publishers, Dor-
drecht, pp. 193-236.
Hickman, M. and C. E. Schweger. 1996. The Late Quater-
nary palaeoenvironmental history of a presently deep
freshwater lake in east-central Alberta, Canada and
palaeoclimate implications. Palaeogeogr., Palaeoclima-
tol., Palaeoecol. 123: 161-178.
Hoffecker, J. F, W. R. Powers, and T. Goebel. 1993. The col-
onization of Beringia and the peopling of the New
World. Science 259: 46-53.
Holloway, R. G., L. M. Raab, and R. Stuckenrath. 1987.
Pollen analysis of Late Holocene sediments from a
central Texas bog. Texas J. Sci. 39: 71-80.
Hostetler, S. W, F Giorgi, G. T. Bates, and P. J. Bartlein. 1994.
Lake-atmosphere feedbacks associated with paleolakes
Bonneville and Lahontan. Science 263: 665-668.
Huntley, B. and T. Webb III (eds.). 1988. Vegetation history.
In: H. Lieth, (ed.). Handbook of Vegetation Science.
Kluwer Academic Publishers, Dordrecht. Vol. 7.
Imbrie, J. and K. P. Imbrie. 1979. Ice Ages, Solving the Mys-
tery. Harvard University Press, Cambridge, MA.
Iversen, J. 1941. Land occupation in Denmark's Stone Age.
A pollen-analytical study of the influence of farmer
culture on the vegetational development. Danmarks
Geol. Unders. 66: 20-65.
Jackson, S. T. and C. R. Givens. 1994. Late Wisconsinan
vegetation and environment of the Tunica Hills region,
Louisiana/Mississippi. Quatern. Res. 41: 316-325.
312 Late Cretaceous and Cenozoic History of North American Vegetation
Jackson, S. T., J. T. Overpeck, T. Webb III, S. E. Keattch, and
K. H. Anderson. 1997. Mapped plant-macrofossil and
pollen records of Late Quaternary vegetation change in
eastern North America. Quaternary Sci. Rev. 16: 1-70.
Jackson, S. T. and D. R. Whitehead. 1991. Holocene vegeta-
tion patterns in the Adirondack Mountains. Ecology
72: 641-653.
. 1993. Pollen and macrofossils from Wisconsinan
interstadial sediments in northeastern Georgia. Qua-
tern. Res. 39: 99-106.
Jacobs, B. E, P L. Fall, and O. K. Davis (eds.). 1985. Late
Quaternary vegetation and climates of the American
southwest. Am. Asso. Stratigr. Palynol. Contrib. Ser.
16:1-185.
Jacobson, G. L., Jr., T. Webb III, and E. C. Grimm. 1987. Pat-
terns and rates of vegetation change during the
deglaciation of eastern North America. In: W E Ruddi-
man and H. E. Wright, Jr. (eds.), North America and
Adjacent Oceans During the Last Deglaciation. The
Geology of North America, Decade of North American
Geology. Geological Society of America, Boulder, CO.
Vol. K-3, pp. 277-288.
Jansen, E. and T. Veum. 1990. Evidence for two-step
deglaciation and its impact on North Atlantic deep-
water circulation. Nature 343: 612-616.
Johnson, C. G., Jr., R. R. Clausmitzer, P. J. Mehringer, and C.
D. Oliver. 1994. Biotic and Abiotic Processes of East-
side Ecosystems: The Effects of Management on Plant
and Community Ecology, and on Stand and Landscape
Vegetation Dynamics. U.S. Dept. Agriculture, Forest
Service, Pacific Northwest Station, Portland, OR. Gen-
eral Tech. Rep. PNW-GTR-322.
Josenhans, H., D. Fedje, R. Pienitz, and J. Southon. 1997.
Early humans and rapidly changing Holocene sea lev-
els in the Queen Charlotte Islands-Hecate Strait,
British Columbia, Canada. Science 277: 71-74.
Kapp, R. O. 1970. Pollen analysis of pre-Wisconsin sedi-
ments from the Great Plains In: W. Dort and J. K. Jones
(eds.), Pleistocene and Recent Environments of the
Central Great Plains. University of Kansas Press,
Lawrence, KS. pp. 143-155.
Kearney, M. S. and B. H. Luckman. 1983. Holocene timber-
line fluctuations in Jasper National Park, Alberta. Sci-
ence 221: 261-263.
Kennett, J. P. and B. L. Ingram. 1995. A 20,000-year record
of ocean circulation and climate change from the
Santa Barbara Basin. Nature 377: 510-514.
Kerr, R.A. 1993. How ice age climate got the shakes. Sci-
ence 260: 890-892.
King, J. E. 1973. Late Pleistocene palynology and biogeog-
raphy of the western Missouri Ozarks. Ecol. Monogr.
43: 539-565.
. 1981. Late Quaternary vegetational history of Illi-
nois. Ecol. Monogr. 51: 43-62.
Klassen, R. W, L. D. Delorme, and R. J. Mott. 1967. Geology
and paleontology of Pleistocene deposits in south-
western Manitoba. Can. J. Earth Sci. 4: 433-477.
Klein, R. G. 1975. The relevance of Old World archeology
to the first entry of man into the New World. Quatern.
Res. 5: 391-394.
Kneller, M. and D. Peteet. 1993. Late-Quaternary climate in
the Ridge and Valley of Virginia, U.S.: changes in veg-
etation and depositional environment. Quatern. Sci.
Rev. 12: 613-628.
Koehler, R A. and R. S. Anderson. 1994. The paleoecology
and stratigraphy of Nichols Meadow, Sierra National
Forest, California, USA. Palaeogeogr., Palaeoclimatol.,
Palaeoecol. 112: 1-17.
Kotilainen, A. T. and N. J. Shackleton. 1995. Rapid climatic
variability in the North Pacific Ocean during the past
95,000 years. Nature 377: 323-326.
Kiichler, A. W. 1964. Manual to Accompany the Map Po-
tential Natural Vegetation of the Conterminous United
States. American Geographical Society, New York,
Special Publ.
Kutzbach, J. E. 1987. Model simulations of the climatic pat-
terns during the deglaciation of North America. In: W.
E Ruddiman and H. E Wright, Jr. (eds.), North America
and Adjacent Oceans during the Last Deglaciation.
The Geology of North America, Decade of North
American Geology. Geological Society of America,
Boulder, CO. Vol. K-3, pp. 425-446.
and R J. Guetter. 1986. The influence of changing or-
bital parameters and surface boundary conditions on
climate simulations for the past 18,000 years. J. Atmos.
Sci. 43: 1726-1759.
Kutzbach, J. E. and H. E. Wright, Jr. 1985. Simulation of the
climate of 18,000 years BP; results for the North Amer-
ican/North Atlantic/European sector and comparison
with the geological record of North America. Quatern.
Sci. Rev. 4:147-187.
Lamb, H. E and M. E. Edwards. 1988. The Arctic. In: B.
Huntley and T. Webb III (eds.), Vegetation History.
Kluwer Academic Publishers, Dordrecht, pp. 519-555.
Larson, D. A., V. M. Bryant, Jr., and T. S. Patty. 1972. Pollen
analysis of a central Texas bog. Am. Midland Nat. 88:
358-367.
Lemons, D. R., M. R. Milligan, and M. A. Chan. 1996. Pale-
oclimatic implications of Late Pleistocene sediment
yield rates for the Bonneville Basin, northern Utah.
Palaeogeogr., Palaeoclimatol., Palaeoecol. 123: 147-
159.
Leopold, E. B. 1956. Two late-glacial deposits in southern
Connecticut. Proc. Natl. Acad. Sci. 52: 863-867.
, R. Nickmann, J. I. Hedges, and J. R. Ertel. 1982.
Pollen and lignin records of Late Quaternary vegeta-
tion, Lake Washington. Science 218: 1305-1307.
Levesque, A. J., L. C. Cwynar, and I. R. Walker. 1997. Ex-
ceptionally steep north-south gradients in lake tem-
peratures during the last deglaciation. Nature 385:
423-426.
Levesque, A. J., E E. Mayle, I. R. Walker, and L. C. Cwynar.
1993. A previously unrecognized late-glacial cold
event in eastern North America. Nature 361: 623-626.
[See Nature 363: 188, for a corrected version of the
pollen diagrams.]
Li, J., T. K. Lowenstein, C. B. Brown, T.-L. Ku, and S. Luo.
1996. A 100 Kya record of water tables and paleocli-
mates from salt cores, Death Valley, California. Palaeo-
geogr., Palaeoclimatol., Palaeoecol. 123: 147-159.
Livingstone, D. A. 1955. Some pollen profiles from Arctic
Alaska. Ecology 36: 587-600.
. 1957. Pollen analysis of a valley fill near Umiat,
Alaska. Am. J. Sci. 255: 254-260.
Love, D. and A. Love. 1974. Origin and evolution of Arctic
and alpine floras. In: J. D. Ives and R. G. Barry (eds.),
Arctic and Alpine Environments. Methuen, London,
pp. 571-603.

Ludlum, D. 1966. Early American Winters: 1604-1820.
American Meterological Society, Boston.
MacDonald, G. M., T. W. D. Edwards, K. A. Moser, R.
Pienitz, and J. E Smol. 1993. Rapid response of treeline
vegetation and lakes to past climate warming. Nature
361: 243-246.
Mack, R. N., N. W. Rutter, and S. Valastro. 1983. Holocene
vegetational history of the Kootenai River Valley, Mon-
tana. Quatern. Res. 20: 177-193.
, and V M. Bryant, Jr. 1978. Late Quaternary vegeta-
tion history at Waits Lake, Colville River Valley, Wash-
ington. Bot. Gaz. 139: 499-506.
Maher, L. J., Jr. 1963. Pollen analyses of surface materials
from the southern San Juan Mountains, Colorado.
Geol. Soc. Am. Bull. 74:1485-1504.
Manabe, S. and A. J. Broccoli. 1985. The influence of con-
tinental ice sheets on the climate of an ice age. J. Geo-
phys. Res. 90: 2167-2190.
Mann, D. H. and T. D. Hamilton. 1995. Late Pleistocene and
Holocene paleoenvironments of the North Pacific
coast. Quatern. Sci. Rev. 14: 449-471.
Markgraf, V, J. P. Bradbury, R. M. Forester, G. Singh, and
R. S. Sternberg. 1984. San Agustin Plains, New Mex-
ico: age and paleoenvironmental potential reassessed.
Quatern. Res. 22: 336-343.
Martin, R S. 1958. Taiga-tundra and the full-glacial period
in Chester County, Pennsylvania. Am. J. Sci. 256:
470-502.
. 1963a. The Last 10,000 Years. A Fossil Pollen
Record of the American Southwest. University of Ari-
zona Press, Tucson, AZ.
. 1963b. Early man in Arizona: the pollen evidence.
Am. Antiq. 29: 67-73.
. 1963c. Geochronology of pluvial Lake Cochise,
southern Arizona. II. Pollen analysis of a 42-meter
core. Ecology 44: 436-444.
-. 1973. The discovery of America. Science 179:
969-974.
Martin, W. C. 1969. Fossil flora of the Rita Blanca lake de-
posits. Geol. Soc. Am. Memoir 113:101-106.
Mathewes, R. W. 1973. A palynological study of postglacial
vegetation changes in the University Research Forest,
southwestern British Columbia. Can. J. Bot. 51: 2085-
2103.
. 1979. A paleoecological analysis of Quadra Sand at
Point Grey, British Columbia, based on indicator
pollen. Can. J. Earth Sci. 16: 847-858.
. 1985. Paleobotanical evidence for climatic change
in southern British Columbia during late-glacial and
Holocene time. Syllogeus 55: 397-422.
. 1991. Climatic conditions in the western and
northern cordillera during the last glaciation: paleo-
ecological evidence. Geograp. Phys. Quatern. 45: 333-
339.
. 1993. Evidence for Younger Dryas-age cooling on
the North Pacific coast of America. Quatern. Sci. Rev.
12: 321-331.
. 1996. Paleoecology of Eraser (Late Wisconsin)
glaciation in coastal British Columbia, Canada. In:
IXth International Palynological Congress, Houston,
TX, June 1996, Program and Abstracts, p. 99.
and J. C. Clague. 1994. Detection of large prehistoric
earthquakes in the Pacific Northwest by microfossil
analysis. Science 264: 688-691.
Quaternary North American Vegetational History 313
Maxwell, J. A. and M. B. Davis. 1972. Pollen evidence of
Pleistocene and Holocene vegetation on the Allegheny
Plateau, Maryland. Quatern. Res. 2: 506-530.
McAndrews, J. H. 1966. Postglacial history of prairie, sa-
vanna, and forest in northwestern Minnesota. Mem.
Torrey Bot. Club. 22: 1-72.
. 1988. Human disturbance of North American
forests and grasslands: the fossil pollen record. In: B.
Huntley and T. Webb III (eds.), Vegetation History.
Kluwer Academic Publishers, Dordrecht, pp. 673-
697.
McCarten, N. and T. R. Van Devender. 1988. Late Wisconsin
vegetation of Robber's Roost in the western Mojave
Desert, California. Madrono 35: 226-237.
McLachlan, J. S. and L. B. Brubaker. 1995. Local and re-
gional vegetation change on the northeastern Olympic
Peninsula during the Holocene. Can. J. Bot. 73:
1618-1627.
Mehringer, P. J., Jr. 1965. Late Pleistocene vegetation in the
Mohave Desert of southern Nevada. J. Ariz. Acad. Sci.
3: 172-188.
. 1967. Pollen analysis of the Tule Springs area,
Nevada. In: H. M. Wormington and D. Ellis (eds.),
Nevada State Museum Anthropological Papers. Carson
City, NV. No. 13. pp. 129-200.
. 1985. Late-Quaternary pollen records from the in-
terior Pacific Northwest and northern Great Basin of
the United States. In: V. M. Bryant, Jr. and R. G. Hol-
loway (eds.), Pollen Records of Late-Quaternary North
American Sediments. American Association of Strati-
graphic Palynologists Foundation, Dallas, TX. pp.
167-189.
-, E. Blinman, and K. L. Petersen. 1977. Pollen influx
and volcanic ash. Science 198: 257-261.
Meltzer, D. J. 1991. Altithermal archaeology and paleoecol-
ogy at Mustang Springs, on the Southern High Plains
of Texas. Am. Antiq. 56: 236-267.
Mikolajewicz, U, T. J. Crowley, A. Schiller, and R. Voss.
1997. Modeling teleconnections between the North At-
lantic and North Pacific during the Younger Dryas. Na-
ture 387: 384-387.
Miller, G. H., J. T. Andrews, and S. K. Short. 1977. The last
interglacial-glacial cycle, Clyde Foreland, Baffin Is-
land, N.W.T.: stratigraphy, biostratigraphy, and
chronology. Can. J. Earth Sci. 14: 2824-2857.
Miller, N. G. 1992. A contribution toward a history of the
Arctic moss flora. Contrib. University of Mich. Herb.
18: 73-86.
. 1993. New Late-Pleistocene moss assemblages
from New England, U.S.A., and their bearing on the
migrational history of the North American moss flora.
J. Hattori Bot. Lab. 74: 235-248.
Miller, R. E and P. E. Wigand. 1994. Holocene changes in
semiarid pinyon-juniper woodlands. BioScience 44:
465-474.
Mills, H. H. and E A. Delcourt. 1991. Quaternary geology of
the Appalachian Highlands and interior low plateaus.
In: R. B. Morrison (ed.), Quaternary Nonglacial Geol-
ogy: Conterminous U.S. The Geology of North Amer-
ica, Decade of North American Geology. Geological So-
ciety of America, Boulder, CO. Vol. K-2, pp. 611-628.
Mix, A. C. 1987. The oxygen-isotope record of glaciation.
In: W. E Ruddiman and H. E. Wright, Jr. (eds.), North
America and Adjacent Oceans during the Last Degla-
314 Late Cretaceous and Cenozoic History of North American Vegetation
ciation. The Geology of North America, Decade of
North American Geology. Geological Society of Amer-
ica, Boulder, CO. Vol. K-3, pp. 111-136.
Mock, C. J. and P. J. Bartlein. 1995. Spatial variability of
Late-Quaternary paleoclimates in the western United
States. Quatern. Res. 44: 425-433.
Moore, E D., J. A. Webb, and M. E. Collinson. 1991. Pollen
Analysis. Blackwell Science Publishers, Oxford, U.K.
2nd edition.
Morgan, A. V. 1987. Late Wisconsin and Early Holocene pa-
leoenvironments of east-central North America based
on assemblages of fossil Coleoptera. In: W. E Ruddi-
man and H. E. Wright, Jr. (eds.), North America and
Adjacent Oceans During the Last Deglaciation. The
Geology of North America, Decade of North American
Geology. Geological Society of America, Boulder, CO.
Vol. K-3, pp. 353-370.
Mott, R. J., T. W. Anderson, and J. V. Matthews, Jr. 1982.
Pollen and macrofossil study of an interglacial deposit
in Nova Scotia. Geograp. Phys. Quatern. 36: 197-208.
Osborn, G. D. 1982. Holocene glacier and climate fluctua-
tions in the southern Canadian Rocky Mountains, a re-
view. Striae 18:15-25.
Overpeck, J., D. Rind, A. Lacis, and R. Healy. 1996. Possible
rate of dust-induced regional warming in abrupt cli-
mate change during the last glacial period. Nature 384:
447-449.
Patterson, W. A. Ill and A. E. Backman. 1988. Fire and dis-
ease history of forests. In: B. Huntley and T. Webb III
(eds.), Vegetation History. Kluwer Academic Publish-
ers, Dordrecht, pp. 603-632.
Payette, S. 1993. The range limit of boreal tree species in
Quebec-Labrador: an ecological and palaeoecological
interpretation. Rev. Palaeobot. Palynol. 79: 7-30.
Peltier, W. R. 1994. Ice age paleotopography. Science 265:
195-201.
. 1995. Response—Paleotopography of glacial-age
ice sheets. Science 267: 536-538. [See also Technical
Comments—Edwards, 1995.]
-. 1996. Mantle viscosity and ice-age ice sheet topog-
raphy. Science 273: 1359-1364.
Peteet, D. M. 1991. Postglacial migration history of lodgepole
pine near Yakutat, Alaska. Can. J. Bot. 69: 786-796.
Peteet, D. M. et al. (+ 5 authors). 1990. Younger Dryas cli-
matic reversal in northeastern USA? AMS ages for an
old problem. Quatern. Res. 33: 219-230.
et al. (+ 5 authors). 1993. Late-glacial pollen, macro-
fossils and fish remains in northeastern U.S.A.—the
Younger Dryas oscillation. Quatern. Sci. Rev. 12: 597-
612.
Pewe, T. L. 1983. The periglacial environment in North
America during Wisconsin time. In: S. C. Porter (ed.),
Late-Quaternary Environments of the United States,
The Late Pleistocene. University of Minnesota Press,
Minneapolis, MN. Vol. 1, pp. 157-189.
Phillips, E M. et al (+ 5 authors). 1996. Chronology for fluc-
tuations in late Pleistocene Sierra Nevada glaciers and
lakes. Science 274: 749-751.
Phleger, E B., E L. Parker, and J. E Peirson. 1953. North At-
lantic foraminifera. Rept. Swed. Deep-Sea Expedit.
1947-1948 7: 1-122.
Porter, S. C. (ed.). 1983. Late-Quaternary Environments of
the United States, The Late Pleistocene. University of
Minnesota Press, Minneapolis, MN. Vol. 1.
Potzger, J. E. 1944. Pollen frequency of Abies and Picea in
peat: a correction on some published records from In-
diana bogs and lakes. Butler University of Bot. Stud. 6:
123-130.
and B. C. Tharp. 1943. Pollen record of Canadian
spruce and fir from Texas bog. Science 98: 584-585.
. 1947. Pollen profile from a Texas bog. Ecology 28:
274-280.
-. 1954. Pollen study of two bogs in Texas. Ecology
35:462-466.
Prentice, I. C. 1988. Records of vegetation in time and
space: the principles of pollen analysis. In: B. Huntley
and T. Webb III (eds.), Vegetation History. Kluwer Aca-
demic Publishers, Dordrecht, pp. 17-42.
Remenda, V. H., J. A. Cherry, and T. W. D. Edwards. 1994.
Isotopic composition of old ground water from Lake
Agassiz: implications for Late Pleistocene climate. Sci-
ence 266: 1975-1978.
Richard, R J. H. 1993. Origine et dynamique postglaciaire
de a foret mixte au Quebec. Rev. Palaeobot. Palynol.
79: 31-68.
. 1994a. Wisconsinan Late-Glacial environmental
change in Quebec: a regional synthesis. J. Quatern. Sci.
9: 165-170.
-. 1994b. Postglacial palaeophytogeography of the
eastern St. Lawrence River watershed and the climatic
signal of the pollen record. Palaeogeogr., Palaeoclima-
tol, Palaeoecol. 109: 137-161.
Richmond, G. M. and D. S. Fullerton. 1986. Quaternary
glaciations in the United States of America. In: V.
Sibrava, D. Q. Bowen, and G. M. Richmond (eds.), Qua-
ternary Glaciations in the Northern Hemisphere. Report
of the International Geological Correlation Programme
Project 24. Pergamon Press, Oxford, U.K. pp. 3-10.
Ritchie, J. C. 1984. Past and Present Vegetation of the Far
Northwest of Canada. University of Toronto Press,
Toronto.
. 1985. Quaternary pollen records from the western
interior and the Arctic of Canada. In: V. M. Bryant, Jr.
and R. G. Holloway (eds.), Pollen Records of Late-Qua-
ternary North American Sediments. American Associ-
ation of Stratigraphic Palynologists Foundation, Dal-
las, TX. pp. 327-352.
-. 1987. Postglacial Vegetation of Canada. Cambridge
University Press, Cambridge, U.K.
Royall, P. D., R A. Delcourt, and H. R. Delcourt. 1991. Late
Quaternary paleoecology and paleoenvironments of
the central Mississippi alluvial valley. Geol. Soc. Am.
Bull. 103: 157-170.
Ruddiman, W. E and H. E. Wright, Jr. (eds.) 1987. North
America and Adjacent Oceans during the Last
Deglaciation. The Geology of North America, Decade
of North American Geology. Geological Society of
America, Boulder, CO. Vol. K-3.
Savage, D. E. and D. E. Russell. 1983. Mammalian Paleofau-
nas of the World. Addison-Wesley, London.
Schoonmaker, P. K. and D. R. Foster. 1991. Some implica-
tions of paleoecology for contemporary ecology. Bot.
Rev. 57: 204-245.
Scuderi, L. A. 1993. A 2000-year tree ring record of annual
temperatures in the Sierra Nevada Mountains. Science
259: 1433-1436.
Shackleton, N. J. 1987. Oxygen isotopes, ice volume and
sea level. Quatern. Sci. Rev. 6:183-190.

and N. D. Opdyke. 1977. Oxygen isotope and
palaeomagnetic evidence for early Northern Hemi-
sphere glaciation. Nature 270: 216-219.
Shackleton, N. J. et al. (+16 authors). 1984. Oxygen isotope
calibration of the onset of ice-rafting and history of
glaciation in the North Atlantic region. Nature 307:
620-623.
Shane, L. C. K. 1975. Palynology and radiocarbon chronol-
ogy of Battaglia Bog, Portage County, Ohio. Ohio J. Sci.
75: 96-102.
. 1987. Late-glacial vegetational and climatic history
of the Alleghney Plateau and the till plains of Ohio
and Indiana, U.S.A. Boreas 16:1-20.
. 1989a. Changing palynological methods and their
role in three successive interpretations of the late-
glacial environments at Bucyrus Bog, Ohio, USA.
Boreas 18: 297-309.
. 1989b. The History of Vegetation, Climate, and
Relict Plant Taxa: Palynology of the Gott Fen/Frame
Lake Area, Portage Co., Ohio. Final Report. Ohio De-
partment of Natural Resources, Columbus, OH. pp.
1-20.
and K. H. Anderson. 1993. Intensity, gradients and
reversals in late glacial environmental change in east-
central North America. Quatern. Sci. Rev. 12: 307-320.
Smith, G. L., J. L. Bischoff, and J. P. Bradbury. 1997. Synthe-
sis of the paleoclimatic record from Owens Lake core
OL-92. Geol. Soc. Am. Sp. Paper 317: 143-160.
Spaulding, W. G. 1990a. Vegetation dynamics during the
last deglaciation, southeastern Great Basin, U.S.A.
Quatern. Res. 33:188-203.
. 1990b. Vegetational and climatic development of
the Mojave Desert: the last glacial maximum to the
present. In: J. L. Betancourt, T. R. Van Devender, and
R S. Martin (eds.), Packrat Middens, The Last 40,000
Years of Biotic Change. University of Arizona Press,
Tucson, AZ. pp. 166-199.
. 1991. Pluvial climatic episodes in North America
and North Africa: types and correlation with global
climate. Palaeogeogr., Palaeoclimatol., Palaeoecol. 84:
217-227.
-, E. B. Leopold, and T. R. Van Devender. 1983. Late
Wisconsin paleoecology of the American southwest. In:
S. C. Porter (ed.), Late-Quaternary Environments of the
United States. The Late Pleistocene. University of Min-
nesota Press, Minneapolis, MN Vol. 1, pp. 259-293.
Stock, C. (revised by J. M. Harris). 1992. Rancho La Brea, a
Record of Pleistocene Life in California. Natural His-
tory Museum of Los Angeles County, Los Angles. 7th
edition, Science Ser. 37.
Stute, M., P Schlosser, J. E Clark, and W. S. Broecker. 1992.
Paleotemperatures in the southwestern United States
derived from noble gases in ground water. Science
256: 1000-1003.
Szeicz, J. M. and G. M. MacDonald. 1991. Postglacial vege-
tation history of oak savanna in southern Ontario. Can.
J. Bot. 69: 1507-1519.
Terasmae, J. 1958. Contributions to Canadian palynology.
Geol. Surv. Can. Bull. 46:1-35.
. 1960. Contributions to Canadian palynology no. 2.
Geol. Surv. Can. Bull. 56: 1-41.
. 1967. A Review of Quaternary Palaeobotany and
Palynology in Canada. Geological Survey of Canada,
Ottawa, Paper 67-13, pp. 1-12.
Quaternary North American Vegetational History 315
, P. J. Webber, and J. T. Andrews. 1966. A study of
Late-Quaternary plant-bearing beds in north-central
Baffin Island, Canada. Arctic 19: 296-318.
Thompson, L. G. et al. (+ 7 authors). 1995. Late glacial stage
and Holocene tropical ice core records from Huas-
caran, Peru. Science 269: 46-51.
Thompson, R. S. 1988. Western North America, vegetation
dynamics in the western United States: modes of re-
sponse to climatic fluctuations. In: B. Huntley and T.
Webb III (eds.), Vegetation History. Kluwer Academic
Publishers, Dordrecht, pp. 415-458.
. 1990. Late Quaternary vegetation and climate in
the Great Basin. In: J. L. Betancourt, T. R. Van Deven-
der, and P. S. Martin (eds.), Packrat Middens: The Last
40,000 Years of Biotic Change. University of Arizona
Press, Tucson, AZ. pp. 200-239.
-, C. Whitlock, P. J. Bartlein, S. R Harrison, and W. G.
Spaulding. 1993. Climatic changes in the western
United States since 18,000 yr B.P. In: H. E. Wright, Jr.
(ed.), Global Climates Since the Last Glacial Maxi-
mum. University of Minnesota Press, Minneapolis,
MN. pp. 468-513.
Thunell, R. C. and P. G. Mortyn. 1995. Glacial climate in-
stability in the northeast Pacific Ocean. Nature 376:
504-506.
Toomey, R. S. Ill, M. D. Blum, and S. Valastro, Jr. 1993. Late
Quaternary climates and environments of the Edwards
Plateau, Texas. Global Planet. Change 7: 299-320.
Traverse, A. 1988. Paleopalynology. Unwin Hyman,
Boston.
Tucker, J. M. 1969. Oak leaves of the Rita Blanca lake de-
posits. Memoir Geol. Soc. Am. 113: 97-99.
Van Devender, T. R. 1985. Climatic cadences and the com-
position of Chihuahuan Desert communities: the Late
Pleistocene packrat midden record. In: J. Diamond and
T. J. Case (eds.), Community Ecology. Harper & Row,
New York. pp. 285-299.
. 1987. Holocene vegetation and climate in the
Puerto Blanco Mountains, southwestern Arizona.
Quatern. Res. 27: 51-72.
. 1990a. Late Quaternary vegetation and climate of
the Chihuahuan Desert, United States and Mexico. In:
J. L. Betancourt, T. R. Van Devender, and P. S. Martin
(eds.), Packrat Middens, The Last 40,000 Years of Bi-
otic Change. University of Arizona Press, Tucson, AZ.
pp. 104-133.
. 1990b. Late Quaternary vegetation and climate of
the Sonoran Desert, United States and Mexico. In: J. L.
Betancourt, T. R. Van Devender, and P S. Martin (eds.),
Packrat Middens, The Last 40,000 years of Biotic
Change. University of Arizona Press, Tucson, AZ. pp.
134-165.
-, J. L. Betancourt, and M. Wimberly. 1984. Biogeo-
graphic implications of a packrat midden sequence
from the Sacramento Mountains, south-central New
Mexico. Quatern. Res. 22: 344-360.
Van Devender, T. R. and D. H. Riskind. 1979. Late Pleis-
tocene and early Holocene plant remains from Hueco
Tanks State Historical Park: the development of a
refugium. Southwest. Nat. 24: 127-140.
Van Devender, T R., R. S. Thompson, and J. L. Betancourt.
1987. Vegetation history of the deserts of southwestern
North America; the nature and timing of the Late Wis-
consin-Holocene transition. In: W. E Ruddiman and
316 Late Cretaceous and Cenozoic History of North American Vegetation
H. E. Wright, Jr. (eds.), North America and Adjacent
Oceans during the Last Deglaciation. The Geology of
North America, Decade of North American Geology.
Geological Society of America, Boulder, CO. Vol. K-3,
pp. 323-352.
VEMAP (Vegetation/Ecosystem Modeling and Analysis
Project). 1995. A comparison of biogeography and bio-
geochemistry models in the context of global climate
change. Global Biogeochemistry Cycles 9: 407-437.
Waddington, J. C. B. and H. E. Wright, Jr. 1974. Late Quater-
nary vegetational changes on the east side of Yellow-
stone Park, Wyoming. Quatern. Res. 4: 175-184.
Walker, H. J. 1973. Morphology of the North Slope. In: M. E.
Britton (ed.), Alaskan Arctic Tundra. Arctic Institute of
North America, Fairbanks, AK, Tech. Paper 25, pp.
49-92.
Walker, I. R., R. J. Mott, and J. P. Smol. 1991. Aller0d-
Younger Dryas lake temperatures from midge fossils in
Atlantic Canada. Science 253:1010-1012.
Watts, W. A. 1970. The full-glacial vegetation of northwest-
ern Georgia. Ecology 51: 17-33.
. 1979. Late Quaternary vegetation of central Ap-
palachia and the New Jersey Coastal Plain. Ecol.
Monogr. 49: 427-469.
. 1983. Vegetational history of the eastern United
States 25,000 to 10,000 years ago. In: S. C. Porter (ed.),
Late-Quaternary Environments of the United States,
The Late Pleistocene. University of Minnesota Press,
Minneapolis, MN. Vol. 1, pp. 294-310.
and B. C. S. Hansen. 1994. Pre-Holocene and
Holocene pollen records of vegetation history from the
Florida Peninsula and their climatic implications.
Palaeogeogr., PalaeoclimatoL, Palaeoecol. 109:163-176.
Watts, W. A. and M. Stuiver. 1980. Late Wisconsin climate
of northern Florida and the origin of species-rich de-
ciduous forest. Science 210: 325-327.
Webb, S. D. 1985. Main pathways of mammalian diversifi-
cation in North America. In: E G. Stehli and S. D. Webb
(eds.), The Great American Biotic Interchange. Plenum,
New York. pp. 201-217.
Webb, T. III. 1987. The appearance and disappearance of
major vegetational assemblages: long-term vegeta-
tional dynamics in eastern North America. Vegetatio
69: 177-187.
. 1988. Eastern North America. In: B. Huntley and T.
Webb III (eds.), Vegetation History. Kluwer Academic
Publishers, Dordrecht, pp. 385-414.
-, P. J. Bartlein, and J. E. Kutzbach. 1987. Climatic
change in eastern North America during the past
18,000 years; comparisons of pollen data with model
results. In: W. E Ruddiman and H. E. Wright, Jr. (eds.),
North America and Adjacent Oceans during the Last
Deglaciation. The Geology of North America, Decade
of North American Geology. Geological Society of
America, Boulder, CO. Vol. K-3, pp. 447-462.
Webb, T. Ill, E. J. Gushing, and H. E. Wright, Jr. 1983.
Holocene changes in the vegetation of the Midwest. In:
H. E. Wright, Jr. (ed.), Late Quaternary Environments of
the United States, The Holocene. University of Min-
nesota Press, Minneapolis, MN. Vol. 2, pp. 142-165.
Whitehead, D. R. 1965. Prehistoric maize in southeastern
Virginia. Science 150: 881-883.
. 1981. Late-Pleistocene vegetational changes in north-
eastern North Carolina. Ecol. Monogr. 51: 451-471.
, D. E Charles, S. T. Jackson, J. R Smol, and D. R. Eng-
strom. 1989. The developmental history of Adiron-
dack (N.Y.) lakes. J. Paleolimnol. 2: 185-206.
Whitehead, D. R. and R. Q. Oaks, Jr. 1979. Developmental
history of the Dismal Swamp. In: P. W. Kirk, Jr. (ed.),
The Great Dismal Swamp. University of Virginia Press,
Charlottesville, VA. pp. 25-43.
Whitehead, D. R. and M. C. Sheehan. 1985. Holocene veg-
etational changes in the Tombigbee River Valley, east-
ern Mississippi. Am. Midland Nat. 113:122-137.
Whitlock, C. 1992. Vegetational and climatic history of the
Pacific Northwest during the last 20,000 years: impli-
cations for understanding present-day biodiversity.
Northwest Environ. J. 8: 5-28.
. 1993. Postglacial vegetation and climate of Grand
Teton and southern Yellowstone National Parks. Ecol.
Monogr. 63: 173-198.
Whitlock, C. and R J. Bartlein. 1997. Vegetation and climate
change in northwest America during the past 125kyr.
Nature 388: 57-61.
Whittaker, R. H. 1970. Communities and Ecosystems.
MacMillan Co., New York.
Whittlesey, C. 1868. Depression of the ocean during the ice
period. Proc. Am. Assoc. Adv. Sci. 16: 92-97.
Williams, M. A. J., D. L. Dunkerley, E De Deckker, A. E Ker-
shaw, and T. Stokes. 1993. Quaternary Environments.
Edward Arnold, London.
Wright, H. E., Jr. (ed.). 1983. Late-Quaternary Environments
of the United States, The Holocene. University of Min-
nesota Press, Minneapolis, MN. Vol. 2.
H. E. Wright Jr., J. E. Kutzbach, T. Webb III, W. E Ruddiman,
S. A. Street-Perrott, and P. J. Bartlein (eds.). 1993.
Global Climates Since the Last Glacial Maximum. Uni-
versity of Minnesota Press, Minneapolis, MN.
Wright, H. E. Jr., A. M. Bent, B. S. Hansen, and L. J. Maher,
Jr. 1973. Present and past vegetation of the Chuska
Mountains, northwestern New Mexico. Geol. Soc. Am.
Bull. 84: 1155-1180.
Wright, H. E. Jr., T. C. Winter, and H. L. Patten. 1963. Two
pollen diagrams from southeastern Minnesota: prob-
lems in the regional late glacial and postglacial vege-
tational history. Geol. Soc. of Am. Bull. 74: 1371-
1396.
Younger Dryas essays. 1993/1995. Quatern. Sci. Rev. 12/14.
General Readings
Bradley, R. S. (ed.). 1991. Global Changes of the Past.
UCARC, Interdisciplinary Earth Studies, Boulder, CO.
Broecker, W. S. 1994. Massive iceberg discharges as triggers
for global climate change. Nature 372: 421-424.
. 1995. The Glacial World According to Wally.
Lamont-Doherty Earth Observatory, Palisades, N.Y.
Chandler, M. 1996. Trees retreat and ice advances. Nature
381:477-478.
Clark, J. S. et al. (+12 authors). 1998. Reid's paradox of
rapid plant migration. BioScience 48: 13-24.
Delcourt, H. R. and P. A. Delcourt. 1991. Quaternary Ecol-
ogy, A Paleoecological Perspective. Chapman & Hall,
New York.
Gibbons, A. 1993. Pleistocene population explosions. Sci-
ence 262: 27-28.
Gibbons, A. 1996. The peopling of the Americas. Science
274: 31-33.

, 1997. Monte Verde: Blessed but not confirmed. Sci-
ence 275: 1256-1257.
Holden, C. 1994. Early American gene clock gains time.
Science 263: 753.
Hostetler, S. W. 1997. Near to the edge of an ice sheet. Na-
ture 385: 393-399.
Keigwin, L. D. 1995. The North Pacific through the millen-
nia. Nature 377: 485-486.
Kerr, R. A. 1991. Ocean mud pins down a million years of
time. Science 254: 802-803.
Lewin, R. 1985. Plant communities resist climatic change.
Science 228:165-166.
Lister, A. M. 1993. Mammoths in miniature. Nature 362:
288-289.
and A. V. Sher. 1995. Ice cores and mammoth ex-
tinction. Nature 378: 23-24.
MacAyeal, D. 1995. Challenging an ice-core paleother-
mometer. Science 270: 444-445.
Meltzer, D. J. 1997. Monte Verde and the Pleistocene Peo-
pling of the Americas. Science 276: 754-755.
Mlot, C. 1995. A high glacier opens a view of the ice age
tropics. Science 269: 32.
Quaternary North American Vegetational History 317
Moore, P. D. 1995. Back to the Little Ice Age. Nature 377:
684-685.
Pastor, J. 1993. Northward march of spruce. Nature 361:
208-209.
Pielou, E. C. 1991. After the Ice Age, The Return of Life to
Glaciated North America. University of Chicago Press,
Chicago.
Potts, R. 1996. Evolution and climate variability. Science
273: 922-923.
Shane, L. C. K. and E. J. Gushing (eds.). 1991. Quaternary
Landscapes. University of Minnesota Press, Min-
neapolis, MN.
Stanley, S. M. 1996. Children of the Ice Age, How a Global
Catastrophe Allowed Humans to Evolve. Harmony,
New York.
Vrba, E. S., G. H. Denton, T. C. Partridge, and L. H. Burckle
(eds.). 1995. Paleoclimate and Evolution, with Empha-
sis on Human Origins. Yale University Press, New
Haven, CT.
Williams, D. K. 1993. The Desert's Past: A Natural Prehis-
tory of the Great Basin. Smithsonian Institute Press,
Washington, D.C.
 NINE
Origins of North American
Biogeographic Affinities
An aspect of plant distribution that has intrigued biogeog-
raphers for over 200 years is the occurrence of similar
biotas in widely separated regions. The North American
flora has affinities with several such areas: the Mediter-
ranean, the dry regions of South America, eastern Asia,
and eastern Mexico. The origin of some patterns is rela-
tively clear, while for others hypotheses are just now
being formulated.
During times when the dogma of permanence of conti-
nents and ocean basins held sway, explanations for these
disjunctions required imaginative thinking that often bor-
dered on the bizarre. The pendulum or schwingpolen hy-
pothesis was offered to explain the perceived bipolar dis-
tribution of several taxa (Gnetum, Magnolia, Pinus section
Taeda; Simroth, 1914). By this view, the Earth swings in
space like a pendulum, creating regular fluctuations in en-
vironments and often causing the symmetrical placement
of taxa at two points on opposite sides of the Earth. Other
disjunctions were explained by casually placing geophysi-
cally impossible land bridges at any point in time between
any two sites where the presence of similar communities
seemed to call for land connections (see review in Simp-
son, 1943). The presence of teeth ofHipparion, an ungulate
related to the horse, in Europe and South Carolina-Florida
prompted French geologist Leonce Joleaud to propose a
land bridge extending from Florida through the Antilles to
North Africa and Spain. Subsequently, to accommodate
eight new passengers, it was broadened to encompass the
entire region from Maryland and Brazil across to France
and Morocco and its life was prolonged to include virtu-
ally all of the Tertiary. With the later discovery that there
were periodicities in similarity between Old World and
New World Cenozoic faunas, the continents were envi-
sioned as moving back and forth like an accordion. George
Gaylord Simpson, who favored the North Atlantic land
bridge to connect North America and Europe, was beside
318
himself with these theories and characterized Joelaud's as
"the climax of all drift theories." The bridge became well
established in the literature even though it never existed in
the Atlantic Ocean (Marvin, 1973). Udvardy (1969) plotted
all the Cretaceous and Tertiary land bridges postulated for
the South Pacific up to 1913. These pathways were fre-
quently unidirectional and taxonomically selective, and
the amount of water displaced would have virtually elimi-
nated all terrestrial habitats (Graham, 1982).
At the other extreme, attempts to determine centers of
origin and explain patterns of distribution were often
overly simplistic and rested on untested assumptions. The
age and area hypothesis suggested that the occurrence of
organisms was simply a reflection of the age of the lineage:
ancient groups had a wider distribution than more recent
ones (Willis, 1922, 1949). A related view was Hulten's
(1937) equiformal progressive areas, whereby organisms
tended to migrate from their site of origin, or from refugia,
in radially symmetrical patterns. Other visions of dispersal
for North American plants and animals were Fernald's
(1925) nunatak hypothesis and Marie-Victorin's (1938)
sidewalk hypothesis. A nunatak is land within the glacial
boundary that was unglaciated or projected above the ice
and served as a refugium for plants and animals during
glacial times. Sidewalks were ice-free corridors within or
along the margin of the ice that connected refugia, and they
served as pathways for interchange and repopulation of
sites after deglaciation. The two latter models are consis-
tent with some distributions, but several of Fernald's nun-
atak areas were later found to be glaciated. Three decades
after Simpson's (1943) frustrations with an essentially nar-
rative biogeography, Nelson (1978) expressed a similar
view and described dispersal biogeography as "a science of
the improbable, the rare, the mysterious, and the miracu-
lous." As will be seen later, the "school" of vicariance bio-
geography has certain mystic qualities of its own.

The geological sciences were not spared imaginative ef-
forts to explain the unknown, including the mechanisms
for continental drift. Eotvos was a "pole fleeing" force
whereby objects on the surface of a rotating sphere were
envisioned as somehow moving laterally in opposite direc-
tions and sliding down the curved surface of the sphere.
The balloon hypothesis was born out of the observation
that if paper representations of the continents were glued
onto a balloon, they would separate and reunite as it ex-
panded and contracted. A mechanism for the required pe-
riodic doubling of the Earth's diameter proved elusive.
All of these early attempts to explain patterns evident in
the modern distribution of organisms and land fragments
failed because of the lack of accurate phylogenies and an
integrated model of plate movement, land bridges, past cli-
mates, and the fossil record. Current distribution patterns
and biotic affinities are residues of history, and their origin
now can be more accurately traced through better docu-
mented, albeit less spectacular, events (Morrone and Crisci,
1995).
FLORISTIC AFFINITIES BETWEEN
SOUTHWESTERN UNITED STATES-NORTHERN
MEXICO AND MEDITERRANEAN REGION
The dry sclerophyllous vegetation of the southwestern
United States continues southward into northern Mexico
and is characterized especially by numerous shared and re-
lated species of Pinus and Quercus. This assemblage is
Origins of North American Biogeographic Affinities 319
Figure 9.1. Juniperus deppeana, Arizona. Juniper also
occurs in the Mediterranean region and is an example of
a Madrean-Tethyan disjunct.
known as Madrean sclerophyllous vegetation after the
Sierra Madre of Mexico. A community similar in structure
and composition occurs in the Mediterranean region and is
called Tethyan sclerophyllous vegetation after the ancient
Tethys Sea whose remnant is represented by the modern
Mediterranean Sea. Examples of the affinity include spe-
cies of Cupressus, Juniperus (Fig. 9.1), Pinus, Aesculus, Ar-
butus, Buxus, Cercis, Datisca, Lavatera, Liquidambar, Pla-
tanus, Prunus, Quercus, Rhamnus, Rhus, Staphylea, and
Styrax. The Madrean-Tethyan hypothesis was proposed to
explain the relationship and suggests that there was a band
of dry sclerophyllous vegetation extending across the mid-
latitudes of North America and Europe in the Early Tertiary
when the continents were closer together (Axelrod, 1975,
fig. 1). Other possible explanations include long-distance
dispersal (Meusel, 1969; Raven, 1971) and convergence
from widespread mesic-habitat boreotropical ancestors as
progressive drying and greater seasonality in rainfall de-
veloped in the southwestern United States-northern Mex-
ico and in the Mediterranean region in Eocene and later
times (Raven, 1973; Wolfe, 1975).
The Madrean-Tethyan hypothesis can be tested by ex-
amining some of the expected consequences if the extant
species were derived from ancestors that were part of a
continuous or nearly continuous belt of Early Tertiary veg-
etation. One consequence is that each group constituting
an example of the Madrean-Tethyan link should be mono-
phyletic (as opposed to polyphyletic, derived indepen-
dently from different ancestral stocks). Another is that be-
cause the marine barrier across the north-central Atlantic
320 Late Cretaceous and Cenozoic History of North American Vegetation
Ocean was extensive after the Early Tertiary, the divergence
of separate Madrean and Tethyan clades should have
begun by the end of the Paleogene. Divergence time can be
estimated from molecular data by isozyme analysis or DNA
sequencing. The time of divergence is calculated from
changes in nucleotide bases, assuming that an accurate rate
of change has been established and that the rate is constant
over long intervals of time (viz., that there is a molecular
clock; see later section). One such study has been made on
a Madrean-Tethyan disjunct (Fritsch, 1996).
The genus Styrax includes S. officinalis subsp. redi-
vivus and subsp. fulvescens, endemic to California (GAL);
S. texanus, S. platanifolius, and S. youngiae from Texas
(TEX) and northern Mexico; and S. officinalis subsp. offic-
inalis native to the eastern Mediterranean region (MED).
Isozyme analysis supported earlier morphological phylo-
genetic conclusions that the Madrean-Tethyan group was
monophyletic, excluding S. jaliscanus and S. americanus,
which were included in Axelrod's (1975, table 3) original
list. This supports the Madrean-Tethyan hypothesis. How-
ever, divergence time was estimated at 5.0 Ma between
MED and GAL and 8.8 Ma between MED and TEX (mean
6.9 Ma) using the method of Nei (1972, 1987). By Sarich's
(1977) method, divergence time between MED and GAL
was 8.5 Ma, and between MED and TEX it was 13.8 Ma
(mean 11.1 Ma). Using the widest range of values (5.0-13.8
Ma), the molecular data still suggest that divergence oc-
curred in the Middle to Late Miocene, when North Amer-
ica and Europe were already well separated, and this is in-
consistent with the hypothesis.
Fossils of Styrax from Europe date from the Early
Miocene (-20 Ma) at localities in France, Germany, and
eastern Europe, which are north of the proposed belt of
sclerophyllous vegetation. Styrax is also known from the
Middlegate flora of west-central Nevada (Chapter 6); but
that flora is early Middle Miocene in age (15.5 Ma), which
is also too young. Other groups cited with Madrean—
Tethyan distribution (Buxus, Quercus, Pinus, Lavatera) are
not monophyletic, and for the genus Datisca chloroplast
DNA restriction site variations suggest a Late Miocene di-
vergence (see Fritsch, 1996 for references).
The principal argument against long-distance dispersal
is that the seeds and fruits of the relevant species are
poorly adapted for transport over great distances. In the
case of Styrax, the seeds are large, rigid, and probably are
not dispersed far from the parent plant. There are no docu-
mented cases of animal dispersal. The seeds are also bitter
or poisonous. Further, the plants are self-incompatible,
which would require the simultaneous introduction to the
same locale of seeds from two compatible individuals.
Convergence of the Madrean and Tethyan species fol-
lowing migration of mesophytic ancestors across the early
Atlantic Ocean is unlikely based on morphological and
isozyme evidence. This would make dry-adapted species
the sister taxa of mesophytic species. In fact, the
Madrean-Tethyan species form one clade and the meso-
phytic eastern North American-eastern Asian species
form another. It is not known whether mesophytic ances-
tors of CAL-MED Styrax migrated across the Bering land
bridge, which was operable in the Late Tertiary, because
there is no fossil record of the genus from that region. Also,
the paleocommunities there do not include dry sclerophyl-
lous types. Thus, although emerging data seem to discount
the Madrean-Tethyan hypothesis, there is presently insuf-
ficient evidence for a viable alternative. Fritsch (1996) sug-
gests that systematic studies similar to those on Styrax be
made on other disjuncts such as Juniperus, Arbutus, Cer-
cis, Platanus, Prunus, and Rhamnus, which all have an ex-
tensive fossil record.
FLORISTIC AFFINITIES BETWEEN WARM
DESERTS OF SOUTHWESTERN UNITED STATES-
NORTHERN MEXICO AND SOUTH AMERICA
The Sonoran-Chihuahuan deserts in the southwestern
United States and Mexico and the monte of Argentina and
the deserts of Chile in South America share some closely
related plants. The nature and origin of this relationship
has been considered in a number of studies (Hunziker et
al., 1972; Kalin Arroyo et al., 1995; Orians and Solbrig,
1977; Raven 1963, 1971, 1973; Raven, and Axelrod, 1975;
Solbrig, 1972, Thrower and Bradbury, 1977). Prominent ex-
amples at the generic level include Larrea tridentata
(North America; Fig. 9.2A.B) and L. divaricata (South
America; creosote bush; possibly conspecific, Hunziker et
al., 1972); Prosopis juliflora (North America) and P. chilen-
sis (South America; mesquite); while Atamisquea emar-
ginata and Koeberlinia spinosa are species common to
both regions. Other examples are Caesalpinia, Celtis, Cer-
cidium, Condalia, Cressa, Cryptantha, Desmanthus, Er-
razurizia, Evolvulus, Fagonia, Flourensia, Gaillardia,
Hedeoma, Hymenoxys, Gilia, Malvastrum, Mentzelia, Poly-
gonum, Proboscidea, Salvia, Schkuhria, Sida, Ziziphus,
and several grasses (Raven, 1963; Solbrig, 1972).
One explanation is that an arid corridor extended be-
tween the two regions (Johnston, 1940; Van Dyke, 1940)
when the continents were joined or were closer together, as
in the Cretaceous and Early Tertiary. There is considerable
evidence against this, including the facts that the associ-
ated faunas are very different and the number of plant ex-
amples is relatively low (< 2% of the flora). If continuous
or nearly continuous land and suitable climates connected
the southwestern United States-northern Mexico with
South America, the similarity would likely be greater.
Among the many examples of desert plants found only in
one of the two regions are Agave, Dasylirion, Eucnide, No-
lina, Olneya, Pachycormus, and Yucca in North America
and Bougainvillea, Bredemeyera, Bulnesia, Monttea, Ram-
orinoa, Trichocereus, and Zuccagnia in South America.
Also, a number of Tertiary palynofloras are known from the
intervening areas of Mexico, Central America, northern

Figure 9.2. Larrea tridentata, Arizona. (Left) Habit. (Right) Close-up showing flowers. The creosote bush also occurs in
the dry regions of Argentina and Chile.
South America, and the Antilles and there is presently no
evidence for a belt of arid vegetation or subhumid environ-
ments (e.g., Graham, 1976, 1985, 1987, 1988a,b, 1989,
1990, 1991a-c; Graham and Jarzen, 1969). Arid habitats
were more extensive in the Late Tertiary than in the Creta-
ceous when maximum elevations were attained in the
coastal mountains, and in the Quaternary when aridity de-
veloped in association with cooling climates. Indeed, dry
areas are more extensive now than at anytime in the Ceno-
zoic. However, in this connection it should be noted that
earlier references to the origin of Mediterranean climates
as dating to the Late Pleistocene (e.g., Raven, 1972; Raven
and Axelrod, 1978) require revision in light of the recent
Greenland ice core and marine oxygen isotope data. Condi-
tions similar to the latest glacial-interglacial cycle have
likely prevailed numerous times since the onset of Late
Cenozoic cold climates in the Late Pliocene (-2.4 Ma) and
Quaternary (Chapter 2, The Greenland Ice Core Record,
The Thermohaline Cell; Chapter 8). In Europe Mediter-
ranean climates date from the Pliocene (Sue, 1984). After
separation of North and South America in the Cretaceous,
a marine barrier extended across much of Central America
until -3.5 Ma; and it was not until -2.4 Ma that land sur-
faces existed for the exchange of large upland terrestrial an-
imals and plants lacking means of long-distance dispersal
(Coates et al, 1992; Graham, 1992). It is likely that the
affinity is of recent origin and is a result of long-distance
dispersal. Many of the plant examples are self-compatible
species, occupy open habitats convenient for introduction
Origins of North American Biogeographic Affinities 321
and establishment, and have edible and/or small seeds.
Each year millions of birds migrate between the regions
(e.g., Charadrius vulgaris, the semipalmated plover; see
Raven, 1972) affording one mode of transport for plant
propagules.
FLORISTIC AFFINITIES BETWEEN EASTERN
NORTH AMERICA AND EASTERN ASIA
The deciduous forest of eastern North America merges
northward through the lake states forest into the boreal
forest and is bordered on the west by the midcontinent
grasslands (Fig. 9.3). Further west and south there is the
low-altitude desert and mid- to high-altitude montane con-
iferous forest where deciduous species are confined mostly
to gallery and mixed gymnosperm-angiosperm transition
vegetation. In western Europe natural arborescent vegeta-
tion is depauperate due to a combination of topography,
climatic events in the Late Cenozoic and Quaternary, and
human history (see later section). In the Middle East, arid
conditions developing in the Late Tertiary reduced and
currently prevent the widespread occurrence of deciduous
forest. The community reappears in climatically and topo-
graphically suitable areas of the Far East, especially in
Japan and in the central provinces of China, where it
shows a remarkable similarity to the vegetation of eastern
North America (Graham, 1972a).
The first recognition that plants similar to those of
322 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 9.3.
Eastern North America-
eastern Asia distribution
of Carya. Reprinted from
Wu (1983) with the permis-
sion of Zhengyi Wu and the
Missouri Botanical Garden.
North America also occur in eastern Asia is recorded in the
dissertation of Jonas P. Halenius (1750; Fig. 9.4), a student
of Linnaeus, entitled Plantae Rariores Camschatcenses
(Graham, 1966, 1972b). Linnaeus sponsored 186 of these
student theses that constitute a series known as the
Amoenitates Academicae. In those days, the procedures
for obtaining advanced degrees differed considerably from
those of today. In Sweden until about 1885 and later in
other European universities, it was the responsibility of the
student to defend in public debate in Latin a thesis for
which the professor was largely responsible. The purposes
of the examination were to demonstrate the student's fami-
larity with the rules of formal disputation and fluency in
Latin. The content of the thesis and its originality did not
matter greatly. Writing about another student's thesis, Lin-
naeus stated, "Mr. Kalm ought to dispute pro gradu de Oe-
conomiae patriae augmento opera L. B. Bjelke. This he
ought to do out of gratitude . . . I will dictate the first ver-
sion." This system differs considerably from the modern
one in which the student is primarily responsible for the
research but is mercifully spared the ordeal of publicly de-
fending it in Latin within the rules of formal debate. It is an
interesting concept, however.1
In paragraph IV of the dissertation, Halenius (Linnaeus)
notes that
Last summer Gregory Demidoff, a distinguished Ukrain-
ian and excellent judge of plants, submitted to the ex-
amination of our President, an enormous collection of
very rare plants which Lerche, a botanist of great in-
sight, has collected toward the end of last year in Kam-
chatka, the remotest part of Asia and that closest to the
American continent. With some astonishment, I saw
among them not only a great number which are found in
Lapland, but others also—some of which were alto-
gether unknown, some only slightly treated before. Fi-
nally I saw some which are also found in Canada, the
reason being that Canada is not far distant from Kam-
chatka. The following are examples of species previ-
ously seen only in North America but now found also at
the farthest limits of Siberia. (1750)
The list is in the polynomial form used prior to Linnaeus'
consistent application of his binomial system and includes
Claytonia, Anemone, Paris, Kleinia, Heuchera, Uvularia,
Spiraea, Asplenium, and Lycopodium.
In a 1787 reprint of Linnaeus' Critica Botanica the list is
given according to the binomial system and Hulten (1927-
1930) cites the names used in Plantae Rariores Camschat-
censes in his synonymmies for the Flora of Kamtchatka.
This makes it possible to relate the names in Halenius's
thesis to those now in use. It is evident from the combined
distribution of the plants that Linnaeus was referring to
eastern North America because this is the region where
they grow together. On the other hand, it is unlikely that
Linnaeus was aware of the major disjunction that the plants
represented because he knew virtually nothing about the
unexplored vegetation of western North America. His
knowledge of the North American flora came from the col-
lections of Clayton in Virginia and those of Sarrazin and
Gaulthier in eastern Canada and he was probably com-
menting more on the widespread distribution of the plants
than on disjunct relationships with the vegetation of east-
ern Asia. That analysis was made by Asa Gray (1840,1846,
1859) and subsequently documented in greater detail by
others (Boufford and Spongberg, 1983; Graham, 1972a; Li,
1952; Tiffney, 1985; Wood, 1970, 1972). Among the tree
and shrub genera illustrating the pattern are Acer, Aescu-
lus, Buckleya, Carya (Fig. 9.3), Celastrus, Cornus, Gleditsia,
Diospyros, Gordonia, Gymnocladus, Hamamelis, Ilex, Liri-
odendron, Magnolia, Mahonia, Nyssa, Pieris, Rhus,
Robinia, Sassafras, Symplocarpus, Symplocos, Vitis, and
Zanthoxylum.
The forests of eastern North America and eastern Asia
have been viewed as remnants of the Arcto-Tertiary ge-
oflora that extended around the temperate latitudes of the
northern hemisphere primarily from the Late Eocene
through about the Middle Miocene and in Europe until the
Early Quaternary. To the extent that the concept implies a
relatively uniform broad-leaved deciduous forest through-
out this zone, it requires modification to include a promi-
nent evergreen gymnosperm element (e.g., Sequoia), to

avoid the impression that its demise in Europe was exclu-
sively a Quaternary event (Chaney, 1944), and that it was a
monolithic assemblage responding to climatic change as a
unified block of vegetation (see Davis, 1983; Wolfe, 1978,
1979). In North America, and especially in Europe, many
temperate deciduous trees were already disappearing in the
Tertiary. The essential point, however, is that even though
the structure, composition, and history of the northern tem-
perate biota have been oversimplified by the geoflora con-
cept, there was a belt of vegetation similar in structure and
composition extending around much of the northern hemi-
sphere during the Tertiary. Present-day Asian genera that
occur in the fossil record of North America include Ailan-
thus (also widespread in the Tertiary of Europe), Amento-
taxus, Cunninghamia, Ginkgo, Glyptostrobus, Metasequoia,
Castanopsis, Cercidiphyllum, Craigia, Cyclocarya, Diplo-
panax, Dipteronia, Engelhardia, Koelreuteria, Mastixia,
Pachysandra, Platycarya, Tapiscia, Trapa, Trochodendron,
and others. Fossil taxa shared between the Paleocene Fort
Union flora and the Altai flora of Xinjiang in northwestern
China include Corylites, Ditaxocladus, Nordenskioldia,
Nyssidium, Paleocarpinus, Platanus, Trochodendroides,
Origins of North American Biogeographic Affinities 323
FIGURE 9.4 TITLE PAGE FR
sertation of Jonas E Halenius, a student of
Linnaeus, where first reference is made to
the similarities in vegetation between
North America and Asia.
and Ziziphoides (Manchester, 1995; Manchester and Guo,
1996). The present-day American genera Chamaecyparis,
Taxodium, Thuja, Torreya, Tsuga, Asimina, Carya, Dio-
spyros, Lindera, Liquidambar, Liriodendron, Magnolia,
Morus, Nyssa, Persia, Robinia, Sabal, Sapindus, and Sas-
safras were in Europe during the late Cenozoic (Niemela
and Mattson, 1966). It was the disruption of various sectors
of this belt at different times during the Cenozoic that pro-
duced the residual floristic similarities evident between
eastern North America and eastern Asia.
An early disruption occurred in the midcontinent re-
gion of North America. Recall that after the retreat of the
Cretaceous sea there was an initial lowering of temperature
and a diversification of Juglandaceae (Carya), along with
the appearance of Ginkgo, Glyptostrobus, Metasequoia,
and genera identified as Alangium, Betula, Carya, Cas-
tanea, Celtis, Cornus, Corylus, Juglans, Magnolia, Ptero-
carya, Quercus, Symplocos, Ulmus, and Zelkova in the
Fort Union flora. In eastern North America pollen of Alnus,
Betula, Carya, and Platycarya is preserved in the Paleocene
Oak Grove core of northern Virginia; in addition, megafos-
sils of Metasequoia, platanoids, and trochodendroids are
 324 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 9.5. Mixed broad-
leaved deciduous and needle-
leaved evergreen forest,
Changpai Mountains, China
(elevation 500-1000 m).
Trees include Acer, Carpinus,
Fraxinus, Juglans, Tilia, and
Ulmus. Reprinted from Hou
(1983) with the permission of
Hsioh-Yu Hou and the Mis-
souri Botanical Garden. »<C #
known from the Atanikerdluk flora of Greenland. The Late
Paleocene-Early Eocene Thyra 0 flora of Greenland in-
cludes Cephalotaxus, Fokienia, Ginkgo, Metasequoia, Be-
tulaceae, Cercidiphyllum, cf. Corylus, Platanus, and others.
In the midcontinent of North America this vegetation was
replaced by a seasonally dry broad-leaved evergreen forest
in the Late Paleocene and Eocene (Willwood flora). Pale-
osol evidence then suggests dry deciduous forest (34 Ma),
dry woodland (33 Ma), wooded grassland with gallery for-
est (32 Ma), and more open savanna with increasing grasses
(30 Ma). The Kilgore flora of northern Nebraska (13-14
Ma) is a streamside assemblage of broad-leaved deciduous
plants with Pinus, Artemisia, and grasses in the surround-
ing uplands. Continued lowering of winter temperatures
and more seasonal distribution of rainfall further disrupted
the deciduous forest and favored savanna and grassland in
the Plains. Further to the west, the rich Miocene forests of
the Columbia Plateau were replaced by expanding mon-
tane coniferous forest, steppe, and desert in the Late
Miocene and especially in the Pliocene with the rise of the
Sierra Nevada and Coast Ranges. Asian exotics (Alangium,
Cyrilla, Trapa, Pterocarya, Zelkova] had mostly disap-
peared from the eastern North American forest by the
Pliocene, persisting somewhat later west of the Rocky
Mountains. The trend toward grassland in the Plains and
desert, steppe, and montane coniferous forest in the west
culminated in the Quaternary with virtual elimination
there of remnants of the deciduous forest. For North Amer-
ica this left a modified version of the earlier formation only
in the eastern part of the continent.
During the Late Tertiary and Quaternary, the diverse
physiography and the north-south orientation of the
American mountains allowed the deciduous forest to per-
sist in local refugia and to migrate in response to changes
in climate. In Europe, however, the vegetation was caught
between the advancing Fennoscandian ice sheet from the
north and the alpine glaciers of the east-west trending
Pyrennes-Alp mountain system to the south. The result
was the disappearance of the gymnosperm-broad-leaved
deciduous forest during glacial intervals. Some elements
probably persisted in refugia on the southern slopes of the
Alps and in the Middle East (van der Hammen et al., 1971);
but like glacial refugia in the southeastern United States,
subsequent reforestation was a complex process with no
single center serving as a central refuge and with various
deciduous forest elements migrating as individuals. Later,
as Holocene forest regeneration was underway, agricultur-
alists began moving in from the Fertile Crescent between
the Tigris and Euphrates Rivers, clearing the land with fire
and ax and planting crops, as reflected in the dramatic
pollen diagrams of Iversen (1941). With the development
of arid conditions in much of the Middle East in the Late
Tertiary and Quaternary, remnants of the once widespread
temperate gymnosperm-broad-leaved angiosperm forest
remain primarily in eastern North America (Fig. 1.9) and
eastern Asia (Fig. 9.5).
FLORISTIC AFFINITIES BETWEEN EASTERN
NORTH AMERICA AND EASTERN MEXICO
The deciduous forest formation extends into northeastern
Texas where a number of temperate genera reach their
southern limits in the conterminous United States (Fig. 9.6;
Chapter 8, Southeast). Farther south are the coastal grass-
lands and interior Chihuahuan Desert of Texas and north-
Figure 9.6. Eastern North America-eastern Mexico distribution of four forest taxa. Reprinted from Martin and
Harrell (1957) with the permission of The Ecological Society of America.
326 Late Cretaceous and Cenozoic History of North American Vegetation
Figure 9.7. Remnants of the deciduous forest, Jalapa,
Veracruz, Mexico. The principal trees are Liquidambar
and Quercus.
ern Mexico. Elements of the temperate broad-leaved decid-
uous forest reappear in a zone between ~1000 and 2000 m
along the eastern escarpment of the Mexican Plateau and
the Sierra Madre Oriental (Gomez-Pompa, 1973; Fig. 9.7).
They extend in diminishing numbers through the state of
Chiapas in southern Mexico (Breedlove, 1973) and into
Guatemala. Pinus reaches its southern limit in the
Cordillera Dariense of Nicaragua (P. oocarpa, P. patula;
Perry, 1991; Styles, 1993); Alnus, Ilex, Myrica, and Quercus
occur scattered through Central America into northern
South America.
Among the plants disjunct between the eastern United
States and eastern Mexico are species of Acer, Alnus,
Carpinus, Carya, Cercis, Cornus, Fagus, Fraxinus, Juglans,
Liquidambar, Magnolia, Myrica, Nyssa, Ostrya, Platanus,
Prunus, Rhus, Smilax, Tilia, and Ulmus. The floristic affin-
ity between the eastern United States and eastern Mexico
has been noted by Fernald (1931), Graham (1973a), Mc-
Vaugh (1943), and especially by Miranda and Sharp (1950).
These and other authors have viewed the plants as Tertiary
relicts, as ultimately they have proved to be. However, this
was initially an intuitive assessment because there were no
paleobotanical data to indicate when northern temperate
elements first became established in Mexico. Recall that
Potzger and Tharp (1943, 1947, 1954) had reported up to
11% pollen of Abies and Picea from central Texas; Brown
(1938) identified megafossils from Quaternary deposits in
Louisiana as northern species of Larix, Picea, and Thuja;
and Wilson (in Davis, 1946) found pollen of Picea in peats
from Florida. These reports seemed to suggest that climates
in the southeastern United States were so cold that the bo-
real coniferous forest extended to the Gulf Coast. The im-
plication was that elements of the deciduous forest were
displaced even farther south into Mexico and that with the
return of warmer conditions in the Holocene, remnants
were left stranded in the midaltitude temperate zones of
eastern Mexico (Deevey, 1949). By this model the origin of
biotic affinities between the two regions dates primarily
from the Pleistocene. Another implication of the model is
that if these climatic changes and migrations characterize
the most recent glacial-interglacial cycle, they must have
happened repeatedly throughout the Quaternary. Earlier it
was thought that there were only four of these cycles
within the last 1.6 m.y. at intervals of -175,000 years. It is
now known from oxygen isotope, Greenland ice core, and
other data that there were 18-20 such cycles; this conjures
up a tempo of disruption and reunion reminiscent of
Joleaud's opening and closing of the Atlantic Ocean.
As noted in Chapter 8, the absence of sufficient informa-
tion from the fossil record of the Gulf Coast and from east-

ern Mexico made it difficult to evaluate the model. How-
ever, other studies on the biogeography and taxonomy of
modern assemblages were creating an intuitive feeling that
modifications would be forthcoming as new data emerged.
For example, the biotic affinities that exist between the
eastern United States and eastern Mexico are mostly at the
generic level, implying a substantial period of geographic
isolation. This is evident in the vegetation (Braun, 1955;
Miranda and Sharp, 1950), reptiles and amphibians (Mar-
tin, 1958), and vertebrates (Martin and Harrell, 1957). If cli-
matic events and the displacement of biotic zones were as
extensive and frequent as suggested by the early paleo-
botanical data, there would be multiple contacts between
the biotas extending up to -11 Kya, which would likely
break down reproductive barriers and maintain similarity
at the species level. There is no evidence in southern Texas
and northeastern Mexico for repeating buried soil horizons
of the podzol type required to support a rich deciduous for-
est, and there is no archeological evidence that Amerin-
dian populations changed cultural activities from those
typical of present-day central Texas to ones suggesting cli-
mates of the upper midwest.
Later, several lines of evidence further documented the
need to modify concepts about the time of origin of floristic
affinities between the eastern United States and eastern
Mexico. One was the discovery that the 11% Abies and
Picea pollen reported from the central Texas bogs was in-
correct and that boreal conifers were represented by -1.5%
of Picea only (Graham and Heimsch, 1960). Another was
that megafossils of the boreal Thuja occidentalis reported
from Louisiana were actually the southern white cedar
Chamaecyparis thyoides (Delcourt and Delcourt, 1977).
These revisions showed that although some Picea filtered
southward, closer to the bogs in central Texas from more
extensive populations in the upper and central midwest,
there was no evidence of a boreal coniferous forest extend-
ing all the way to the Gulf Coast. This further reduced the
likelihood that the eastern Mexican temperate elements
were exclusively a result of repeated and frequent inter-
change during the Quaternary.
Evidence that north temperate deciduous forest species
were already established in eastern Mexico in the Tertiary
came from a study of pollen and spores from the Paraje
Solo Formation near Coatzacoalcos, Veracruz, Mexico (Gra-
ham, 1976). These deposits were initially estimated to be
Miocene in age, but ostracode and nannofossil assemblages
now show they are of Middle Pliocene age (-3.4 Ma).
Among the plant microfossils recovered were Abies, Picea,
Alnus, Celtis, Ilex, Juglans, Liquidambar, Populus, Quer-
cus, and Ulmus. It has been suggested that these represen-
tatives of the deciduous forest were not introduced as a re-
sult of Late Cenozoic cooling (Graham, 1973b) but were
remnants of a more ancient widespread vegetation existing
from Paleogene times (Axelrod, 1975). Initially the paleo-
botanical data showed only that none of these elements
were present in Eocene or Miocene deposits of northern
Origins of North American Biogeographic Affinities 327
Latin America (Graham, 1985 et seq.) but that they were
present in the Eocene and later epochs of the eastern
United States (Frederiksen, 1981; Gray, 1960), suggesting a
southward introduction with cooling Cenozoic climates. A
new study of plant microfossils from the Simojovel Group
of Chiapas, Mexico (A. Graham, unpublished data), further
shows that northern temperate elements were not yet es-
tablished in that region in the Oligo-Miocene. Although
the data base for northern Latin America is still meager, if
elements of the eastern deciduous forest were present there
in Early and Middle Tertiary times, these phantom forests
left no evidence in the fossil floras known to date. As noted
long ago by Arnold with reference to Arber and Parkin's
(1907) hypothetical Hemiangiospermae, "workable phylo-
genies [and paleocommunities] cannot be built upon forms
that exist only in the mind." (1947). The paleotemperature
curve (Fig. 3.1) and the existing paleobotanical evidence
suggests that the initial affinities probably date from the
Middle Miocene when global temperatures reached new
lows, allowing the deciduous forest to reach its southern-
most extent. Later Cenozoic and glacial-interglacial arid-
ity left elements isolated in the midaltitude temperate en-
vironments of eastern Mexico. The question is still open as
to whether some elements were reintroduced after the
Miocene through climatic change and by long-distance
transport. This possibility is currently being addressed
through new and innovative biogeograpic analyses.
EMERGING METHODS IN BIOGEOGRAPHY
An essential prerequisite for determining the origin of bio-
geographic affinities is an accurate phylogeny. Taxonomic
treatments that do not closely reflect phylogenetic relation-
ships may suggest false patterns and affinities by unifying
taxonomically heterogeneous assemblages or dividing nat-
ural assemblages into taxa distinguished primarily by ge-
ography. As noted by Raven:
The moss Macromitrium sullivantii C. Mull., thought to
be an endemic of a small area of the southeastern Blue
Ridge escarpment of the United States, has recently
been shown to be identical with Macrocoma hymenos-
tomum (Mont.) Grout, a well known species that occurs
throughout the American tropics and South America.
The range as now understood becomes a striking exam-
ple of a disjunct distribution. On the other hand, the
genus Boisduvalia (Onagraceae), comprising six species
of semiarid western North and South America, was
until recently thought to include a species of the moun-
tains of southeastern Australia and Tasmania, B. tas-
manica (Hook, f.) Munz. With the demonstration that
this species is actually an Epilobium, now known as E.
curtisiae Raven, which forms natural hybrid populations
with closely related Australian species, the situation de-
mands a very different interpretation. (1972, p. 234)
The realization has emerged, especially within the past
decade, that morphological features alone are often insuf-
328 Late Cretaceous and Cenozoic History of North American Vegetation
ficient to reveal phylogenetic relationships. Powerful new
data sources and analytical tools are now available to com-
plement traditional morphological approaches. In addition
to clarifying natural relationships, these techniques are
being applied to questions of biogeography. As noted by
Bremer, "It is the previous methodological approaches to
the search for centers of origin [and patterns of subsequent
dispersal], not the search per se, that are spurious." (1992).
The new methods are being refined and are yielding results
of potentially great value to biogeography.
Allozyme-lsozyme Analysis and
Genetic Distance
It has been known for some time that within individual or-
ganisms there are multiple forms of enzymes that catalyze
the same reaction (Hunter and Markert, 1957; Markert and
M011er, 1959). Earlier, when the genetic basis for these dif-
ferent forms was unknown, they were collectively referred
to as isozymes. Now that genetic factors determining some
of them are better understood, a more precise terminology
is emerging. Different forms of an enzyme encoded by dif-
ferent alleles at the same gene locus are called allozymes.
Forms catalyzing the same reaction encoded at different
loci, or when their genetic basis is unknown, are called
isozymes (Crawford, 1990). The enzymes may be separated
from leaves, seeds, and other plant tissue by gel elec-
trophoresis and bands on the gel can be compared within
populations of the same species and between different
species. The data may be analyzed via genetic identity or
genetic distance calculations (Nei, 1972), and a value is ob-
tained that may reflect the degree of relationship and rela-
tive times of divergence. The method has been used to as-
sess the relationship between populations and species of
Liquidambar disjunct in eastern North America, eastern
Asia, and eastern Mexico.
Although the species of Liquidambar are morphologi-
cally similar, the level of isozyme divergence is compara-
tively high (Hoey and Parks, 1991, 1994). Liquidambar
styraciflua in the eastern United States shows little diver-
gence among different populations, while those in eastern
Mexico exhibit greater divergence. This may be the result
of range restriction into the various refugia described for
eastern North American during glacial intervals so that
most of its current range is of recent origin. If populations
in eastern Mexico were not subjected to the same homoge-
nizing effect of Quaternary climatic change, they may have
experienced less alteration in range and persisted as a
more continuous community for longer periods of time.
Another intriguing possibility raised by the isozyme
studies is that there may have been contact between the
populations of L. styraciflua in the eastern United States
and eastern Mexico after their initial introduction into
Mexico in the Tertiary. There is high genetic identity
among three populations from Mexico and those from the
eastern United States (0.985, 0.930, 0.940). These values
are greater than would be expected if the populations had
been isolated since the Middle to Late Miocene (see dis-
cussion below). This supports the possibility of some inter-
change through climatic fluctuations during the late Ceno-
zoic and Quaternary and throughout the Cenozoic by
long-distance dispersal.
A further example of the potential value of isozyme
studies for better understanding North American floristic
affinities is provided by comparison of populations of Liq-
uidambar in eastern North America and eastern Asia. Ge-
netic identity calculations between the eastern American
L. styraciflua and the Asian L. orientalis, L. formosana, and
L. acalcina gave values of 0.512, 0.431, and 0.485. This
suggests a longer period of isolation than between the east-
ern North American-eastern Mexican populations, dating
perhaps also from the Miocene but without subsequent in-
terchange. Similar molecular divergence analyses have
been made on the Magnolia section Rytidospermum (Qiu
et al., 1995a). When molecular data for the taxa were com-
pared, several levels of divergence were evident among the
eastern Asia-eastern North America disjuncts. This is con-
sistent with the emerging view that organisms migrate as
individuals rather than as unified blocks of vegetation; that
is, various genera and species within a genus may respond
differently and follow individual routes of migration in re-
sponse to changes in climate and the physical environ-
ment, encounter different barriers over time, and be selec-
tively affected by epidemic disease. All of these data are
preliminary, but they clearly demonstrate the usefulness of
molecular information in better quantifying taxonomic re-
lationships and in establishing the time of origin of floristic
affinities.
Molecular Clocks
Another early observation that has great potential for phy-
logenetic and biogeographic studies was that the rate of
amino acid substitution in hemoglobin and cytochrome C
proteins is about the same among different mammalian
groups (Margoliash, 1963; Zuckerkandl, 1987; Zucker-
kandl and Pauling, 1962, 1965a). Later Zuckerkandl and
Pauling (1965b) proposed that the rate of protein change
(molecular evolution) is constant over time in all lin-
eages—that is, there is a molecular clock (Li and Graur,
1991). The method is relevant to biogeographic problems
of the North American flora because, for example, it has
been suggested that the Mexican populations of Liquidam-
bar (L. macrophylla, Gomez-Pompa, 1973) represent a
separate species from that in eastern North America (L.
styraciflua). If so, the protein differences could be used to
date the time of species divergence that would also reflect
the time of geographic isolation.
Unfortunately, there are uncertainties in the method
that need to be resolved. One is that the uneven pace of
speciation reflected by the fossil record is not in accord
with the constant rate implied by the molecular clock.

Figure 9.8. Steps in the preparation of area cladograms from phylogenetic and historical geologic data. (A,D) A, Eurasia;
B, North America; C, Australia; D, South America; and E, Africa. Adapted from Brooks and McLennan (1991, chapter 7).
Modern genetic theory also suggests that the tempo of evo-
lution varies among different groups of organisms and that
it is much more rapid during times of adaptive radiation, at
the margin of ranges where the taxon is at the limits of its
ecological tolerance, and during times of increased envi-
ronmental instability. Other problems include different
generation times for various lineages and the differing effi-
ciency of DNA repair systems. Thus, it has yet to be estab-
lished that there is a molecular clock and that it runs at a
constant pace. Even so, it is worthwhile for biogeographers
to follow developments in this field because refinements
could make available an additional technique for investi-
gating the temporal history and relationships of elements
in the North American flora.
Cladistics, Area Cladograms, and
Vicariance Biogeography
Another powerful tool for estimating natural relationships
among taxa is cladistic analysis (Cronquist, 1987; Donog-
hue and Cantino, 1988; Humphries and Chappill, 1988;
Ronquist, 1997 Wiley et al., 1991). Morphological, molec-
ular, and cytological characteristics of taxa may be entered
into the rows and columns of a data matrix. The character
states are coded in multistate or binary format (e.g., 0 = fea-
Origins of North American Biogeographic Affinities 329
ture present, 1 = feature absent). The information is sub-
jected to cladistic analysis via software such as PAUP or
Hennig86. These programs include facilities for polarizing
characters by outgroup comparison and weighting of char-
acters when one (e.g., fusion of petals) is considered more
important than others (e.g., petal color). The programs gen-
erate a series of trees or cladograms showing branching
patterns from which ancestors, divergence, and evolution-
ary patterns can be inferred and which cluster taxa on the
basis of shared derived characters into groups or "clades"
(Fig. 9.8A). Several to many trees are usually produced and
the selection of the one(s) most likely reflecting the "true"
phylogeny is based on the principle of parsimony: in the
absence of evidence to the contrary, the shortest tree is fa-
vored (i.e., the one invoking the fewest number of steps in
the development of each character). Another option is the
generation of a consensus tree when there are several
equally parsimonious trees. From these cladograms phylo-
genies can be inferred and monophyletic or "natural"
groups (clades) can be identified.
Cladograms are important in biogeography for two prin-
cipal reasons. One is that they can provide an improved
understanding of natural relationships among organisms
that allows real rather than imaginary geographic affinities
to be deduced (Qiu et al., 1995b). The other is that with the
330 Late Cretaceous and Cenozoic History of North American Vegetation
taxon cladogram as a basis, another type of cladogram can
be generated by using distributions and events in historical
geology from which centers of origin can be inferred (e.g.,
Bremer, 1992) and which can suggest the origin of some
modern patterns of distribution. These are called area
cladograms, and this emerging method is based on a syn-
thesis of primarily plate tectonic data and phylogenetic
systematics (Nelson and Platnick, 1981). Their construc-
tion is described by Brooks and McLennan (1991) and Page
(1988).
Assume that a phylogenetic cladogram has been gener-
ated for five taxa (Fig. 9.8A). Vicariance events within the
present range of the collective taxa are identified. In this
example, it is the breakup of Pangaea into Gondwana and
Laurasia (Fig. 9.8B). Other possibilities could be the sepa-
ration of India from Africa, the periodic isolation of Eng-
land from continental Europe as a result of Quaternary sea-
level changes, or the rise of the Rocky Mountains dividing
once continuous ranges into eastern and western popula-
tions. The taxa are assigned numbers and a list is made of
all areas in which they are found:
A = Eurasia sp. 1 = 1
B = North America sp. 2 = 2
C = Australia sp. 3 = 3
D = South America sp. 4 = 4
E = Africa sp. 5 = 5
The taxa numbers are placed on the cladogram and each
internal branch is numbered (Fig. 9.8C). These numerical
codes for the taxa and areas are then entered into a data
matrix using 1 as present and 0 as absent:
Taxon Branches
1 2 3 4 5 6 7 8 9
Binary Code
sp. 1 100001001
sp. 2 010001001
sp. 3 001000011
sp. 4 000100111
sp. 5 000010111
The species names are then replaced with their geo-
graphic distribution:
A Eurasia 100001001
B North America 010001001
C Australia 001000011
D South America 000100111
E Africa 000010111
An area cladogram can now be generated that shows the
evolution of the group in relation to the history of the geo-
graphic areas it occupies (Fig. 9.8D). A comparison of Fig.
9.8A (the phylogenetic cladogram) and Fig. 9.8D (the area
cladogram) shows complete correspondence between clades
and geography. In this simplified example, speciation events
correspond perfectly to past geologic history and the infer-
ence is that speciation in this group was a consequence of
the breakup of Pangaea into Laurasia (spp. 1, 2) and Gond-
wana (spp. 3-5). The relative chronology of these vicari-
ance events can also be plotted on the area cladogram.
There are more direct ways to construct area cladograms
(Forey et al., 1994; Humphries and Parenti, 1986), but the
example used here provides a more stepwise description.
In actual cases there is usually less than 100% correspon-
dence between phylogenetic and area cladograms, indicat-
ing that some process (es) other than vicariance (movement
of land) is operating (e.g., movement of organisms or disper-
sal, and extinctions). However, in ideal situations area
cladograms may distinguish between vicariance and disper-
sal or other events. When different taxa show the same re-
peated patterns of geographical speciation, it strengthens
the proposition that the vicariance event (e.g., the breakup
of Pangaea), rather than dispersal, was primary in the phy-
logeny and in the modern distribution of the clade. In
many regions, such as the Caribbean, the geologic history
is problematic. Here area cladograms eventually may prove
useful in suggesting previously unrecognized tectonic events
made evident by biogeographic analysis. There is no con-
sensus on this point, and indeed the whole subject of vic-
ariance biogeography and the practical (as opposed to the-
oretical) value of area cladograms is still being debated
(Hedges et al., 1994; Page and Lydeard, 1994).
For Canada and the conterminous United States the vic-
ariance approach has been used to study the distribution
and speciation of fish faunas, correlation of salamander
ranges and suspect terranes in the western United States,
and mammal occurrences in the montane regions of west-
ern North America. A number of animal groups have also
been analyzed from the southeastern United States (pocket
gopher, deer mouse, sunfish, salamander; Avise et al.,
1987). Area cladograms for the grasshopper mouse genus
Onychomys in arid regions of western North America sug-
gest that its distribution reflects the pluvial-interpluvial
phases that fragmented arid habitats (Riddle and Honey-
cutt, 1990). Among plants, analyses of phylogenies and
vicariance events have been made on ferns and angio-
sperms in Europe (see Wiley, 1988). However, for North
America examples from plants are rare, mostly dealing
with geographic affinities with distant regions. Vicariance
cladograms have been used to suggest that warm temperate
species of Crataegus crossed Beringia from North America
in the Tertiary and migrated southward into eastern Asia
(Phipps, 1983). Modern biogeographic studies have been
made on the biota of the Hawaiian Islands (Wagner and
Funk, 1995), which provides an ideal setting for vicariance
biogeographic analyses. In the conterminous United States,
however, separation of land fragments by plate tectonic or
orogenic mechanisms were either minor, gradual, occurred
so long ago, and/or have been overprinted by dispersal, ex-
tinctions, and other factors that area cladograms have not
been used for explaining distribution-speciation patterns
for plant taxa in this continental setting.
Cladistic methods assume hierarchical rather than retic-

ulate patterns of evolution (Rieseberg, 1995). In other
words, cladistics does not handle reticulation (viz., hy-
bridization) well. As an extrapolation, it is also true that
vicariance biogeography does not handle the collision of
land fragments bearing separate biotas well. This has oc-
curred among the islands of the Caribbean. For example, in
the Late Miocene Haiti and the Dominican Republic fused
into the present-day island of the Dominican Republic.
The suture is the Cul-de-Sac and Enriquillo Basin north of
Port-au-Prince (Graham, 1990). Such events have not ex-
tensively affected the biota of North America, although
there are exceptions like the transported Mint Canyon and
Carmel floras of California and the Yakutat block of Alaska.
Although there are obvious limitations to the use of vic-
ariance biogeography and area cladograms, the potential is
there for important contributions to our understanding of
distribution patterns. Through continued reexamination of
assumptions and refinement in procedures, area clado-
grams that have proven useful in identifying centers of ori-
gin and pockets of speciation and endemism may suggest
probable tectonic-orogenic events, geographic affinities,
and pathways of migration for elements of the North Amer-
ican flora. In addition, a by-product of these emerging
methods is that they have revived interest in biogeography
and are attracting a new generation of investigators in-
trigued by molecular, mathematical, and theoretical con-
siderations; this is to the benefit of the science.
EPILOGUE
This survey of North American vegetation and paleoenvi-
ronments began with a Late Cretaceous continent having
rugged mountains in the east and modest topographic re-
lief in the west. It was divided into eastern and western
provinces by an epicontinental sea that extended from the
Gulf of Mexico to the Arctic Ocean. Sea level was higher by
300 m, land area was 40% less, and temperatures were
10—12° C higher than at the last glacial maximum. The veg-
etation consisted of four formations: tropical forest, para-
tropical rain forest, notophyllous broad-leaved evergreen
forest, and polar broad-leaved evergreen forest. Initially
gymnosperms formed the canopy in many areas; various
herbaceous-shrubby dicots, palms, ferns, and bryophytes
formed the ground cover. Angiosperms were more promi-
nent in disturbed areas, as along streams, where the unpre-
dictability of the habitat contributed to the development of
deciduousness.
Increased warmth and precipitation in the Paleocene
and Early Eocene resulted in a tropical rain forest, para-
tropical rain forest, notophyllous broad-leaved evergreen
forest, and a much restricted polar broad-leaved deciduous
forest in the uplands of the high latitudes.
Cooling in the Middle Eocene produced intermittent
glaciers on Antarctica and the beginning of a montane
coniferous forest in the volcanic highlands west of the
Origins of North American Biogeographic Affinities 331
Rocky Mountains. In the Oligocene global climates fluctu-
ated between cool and warm, modified locally by topogra-
phy. Glaciers gradually became a more widespread and
consistent feature of the Antarctic landscape and increas-
ingly contributed cold water to circulating ocean currents.
Elements of the broad-leaved deciduous forest and the
montane coniferous forest consolidated into communities
resembling the modern formations. The montane conifer-
ous forest moved into lower elevations to form the boreal
coniferous forest. The broad-leaved deciduous forest
spread southward and joined elements from other locales
in a complex pattern of evolution and migration that has
been oversimplified by the geoflora concept. Another drop
in temperature in the Middle Miocene ushered in Arctic
glaciation, cooler temperatures, and more seasonal rain-
fall. Preadapted local Eocene progenetors of shrubland/
chaparral-woodland-savanna consolidated, the broad-
leaved deciduous forest extended as far south as the up-
lands of eastern Mexico, and trees began to decrease in the
central and northern plains and in the high latitudes. By
the Late Pliocene temperatures were even cooler and rain-
fall was less and more seasonal. North American vegeta-
tion then consisted of near tundra, coniferous forest, decid-
uous forest, near grasslands, near deserts, shrubland/
chapparal-woodland-savanna, and a few residual tropi-
cal elements in the southeastern United States. After 1.6
Ma the modern plant formations had developed, but they
were characterized by associations of a composition and
range that fluctuated with the rapid environmental
changes of the Quaternary. For most of the past 1.6 m.y.,
which was glacial in nature, the prominent vegetation for-
mations of North America were widespread coniferous for-
est (boreal and montane); a deciduous forest restricted
to local refugia; grassland; woodland (oak-hickory, jumper-
pinon pine) in warmer regions, spruce savanna in the
upper midwest, and Artemisia steppe in cooler western re-
gions; elements of a desert vegetation restricted to favor-
able slope and edaphic habitats in the pluvial southwest-
ern United States; and expanded Arctic and alpine tundra.
The present interglacial represents conditions of anom-
alous warmth with widespread deciduous forest in the
east, desert in the west, and tropical elements in the south-
eastern United States.
The recent decade has witnessed many new and im-
proved methodologies and numerous contributions of fun-
damental importance to our understanding of North Amer-
ican Late Cretaceous and Cenozoic environments. There
has been recognition from the study of ocean and Green-
land-Antarctic ice cores that great and sudden climatic
changes characterize recent geologic time, especially at
glacial-interglacial transitions. It seems increasingly likely
that the forcing mechanisms for these changes are more
global in extent than suggested by earlier models based pri-
marily on data from the North Atlantic. The new data in-
clude evidence for cool intervals in Florida and the Pacific
Northwest, which were identified from pollen and spore
332 Late Cretaceous and Cenozoic History of North American Vegetation
diagrams, that correlate with Heinrich events. Some dia-
grams from the Pacific Northwest also provide preliminary
evidence for a cold period corresponding to the Younger
Dryas. The view is emerging that fluctuations and cooling
in the tropics (Barbados, Peru) were greater than suggested
by the earlier CLIMAP studies. Glaciers were present on
Antarctica in the Eocene, rather than first appearing in the
Miocene; the real extent of exotic terranes in the American
west is being documented; and a greater role for solar vari-
ablity is emerging as a potential forcing mechanism for cli-
matic change.
Paleopalynological and paleobotanical studies are also
contributing other new information that is modernizing our
concepts of North American vegetational history. Detailed
paleofloristic studies and revisions are underway on floras
from the Eocene of the southeastern United States; the Pa-
leocene and Eocene of the Plains states and adjacent north-
ern Rocky Mountains; and in Oregon, Washington, and Al-
berta. Additions continue to be made to Miocene floras
from Vermont and Oregon, and a better understanding is
emerging of megafossil and microfossil floras from the high
latitudes. These revisions allow paleomonographic studies
of lineages, which are facilitated by increasing international
cooperative efforts. Such efforts are witnessed by the recent
Workshop on US-Eastern European Collaboration in Paleo-
botany organized by Steven Manchester and sponsored by
the National Science Foundation as part of the International
Organization of Paleobotany meetings in Santa Barbara,
California, in 1996. Other cooperative projects were initi-
ated or further developed as part of the International Con-
ference on Diversification and Evolution of Terrestrial
Plants in Geological Time in Nanjing in 1995, and there are
numerous other examples. This trend toward globalization
of paleobiological research, and its integration into other
types of large-scale environmental studies, is an exciting
development and it is essential for achieving a full and ac-
curate understanding of biotic history.
As large amounts of data accumulate, it has become
necessary to communicate the results electronically and to
develop data bases to store and manipulate information.
Examples can be located on the Worldwide Web via search
engines under such topics as paleobotany, palynology, pa-
leoclimatology, and related subjects. They include the
World Data Center-A (http://www.ngdc.noaa.gov/paleo/
paleo.html), the bulletin board POLPAL-L (listserv®
uoguelph.ca), the commercial Palynodata Inc. (CD-ROM
disks by subscription), the home page of the American As-
sociation of Stratigraphic Palynologists, the list service PA-
LEOLIM (paleolim@nervm.nerdc.ufl.edu), the Plant Fossil
Record Data Base, Electronica Paleontologica (http://www.
nhm.ac.uk/paleonet/pe/glines.html) and many others.
Detailed and high-resolution studies of individual fossil
floras are revealing the effects of volcanism on floras, fau-
nas, and their representation in the fossil record. These
studies further reveal that abrupt threshold changes prob-
ably characterized climates and biotas throughout Late
Cretaceous and Cenozoic time, as shown by the recently
identified cool interval within the Paleogene period of
maximum warmth. Paleosols, otoliths, phytoliths, packrat
middens, dendrochronology, isotopes, seismic profiles,
and stomatal analyses are further expanding the arsenal of
investigative techniques.
There has been a gradual, hard-fought acceptance that
the idea of geofloras requires at least substantial renova-
tion, with the concomitant emergence of the boreotropical
concept as an alternative for envisioning the development
and movement of high-latitide biotas through time. The
temporal nature of associations and the increasing docu-
mentation that organisms migrate as individuals requires
rethinking not only of geofloras, but also of time-honored
views about climax communities. History has repeatedly
shown, however, that new ideas can generate exuberant
claims of success, as well as entrenched attachment to old
ways and previous positions. Eventually a rational balance
is usually achieved. In the case of community stability ver-
sus randomness, climax, and geofloras, it must also be ac-
knowledged that some assemblages do exist in the Tertiary
that resemble modern associations and that more and
longer lasting ones are evident at the level of formations.
Equally vigorous debates have centered around the use
of leaf physiognomy as a means of reconstructing past cli-
mates, but continued refinements and pioneering innova-
tions such as CLAMP and associated statistical analyses
are providing convincing successes for the method. In
turn, this has generated new interest and opportunities for
estimating paleoelevations. If local paleoelevations near
modern values were present in western North America by
the Paleogene, this would be of great importance for
models attempting to simulate the effects of uplift on at-
mospheric circulation and climate. The challenges of in-
corporating molecular technologies and cladistics into
biogeography and phylogenetic reconstructions are attract-
ing new investigators into this rejuvenated field. Alke-
nones, hopanes, and sterenes in sediments, oxygen and hy-
drogen isotopes in wood, flavonoids in fossil leaves of
Zelkova, and ancient DNA in fossil tissues are related dis-
coveries. As noted in the Prologue, there is an excitement
attendant to the realization that with the development and
application of new technology, we are on the threshold of
attaining a much improved understanding of the dynamics
of past environmental change and biotic response. The im-
portance of this understanding for predicting future
changes is just being realized.
Prediction is very difficult—especially when it involves
the future. (Nils Bohr, quoted from Wehr, 1995)
The future of North American plant formations is unclear
because of the novel forcing mechanisms of greenhouse
warming, ozone depletion, destruction of the rain forest,
and unprecedented environmental disruption through in-
tensifying agricultural and industrial activities. The next
generation of paleoecologists will witness clearer signals

about trends in climate, biotic response, relevancy of past
patterns to future events, and the degree to which computer
modeling can predict these events. The task will be compli-
cated, however, by continuing rapid modification of the
modern analogs used to interpret the ancient communities.
After all, as Derek Ager notes, "It's the only present we've
got." (1993, chapter 12).
Beyond theoretical considerations, the world's plant re-
sources are of immense economic, as well as strategic, im-
portance: witness the increasing use of food as an induce-
ment for political conformity. Clearly shifts in climate
affect the geography of agricultural surplus and deficit
(Nicholls, 1997) and some could have serious political im-
plications. Computer models predict that greenhouse
warming by enhanced CO2 concentration will benefit agri-
culture in North America and adversely affect it in the
lower latitudes that include many developing countries
that can least afford a reduction in agricultural productiv-
ity (Rosenzweig and Hillel, 1995). Industrialized countries
will certainly not be immune to the unrest associated with
this reduction.
Predicting the impact of future climatic changes on sea
levels is critical for coastal centers of population and for de-
termining suitable sites for the storage of nuclear waste
with a half-life up to 10-30 ka (Leopold, 1987). The recent
collapse of an underground salt dome in Louisiana illus-
trates the vulnerability of such potential sites to the in-
evitable ebb and flow of groundwater and dissolution re-
sulting from sea-level changes. The recent revelations that
glacial-interglacial stages, rather than proceeding at a
steady pace of 175 Ky, consist of interglacials lasting only
10-12 Ky and that the present one began -11 Kya, under-
scores the importance of paleoclimatic data in formulating
rational long-term strategies and resource management
practices.
Vegetational history provides important proxy data for
determining past patterns of environmental change. It is
also our principal measure of how far we have gone in cre-
ating a future ecological catastrophy through the destruc-
tion of rain forest and other habitats and through unprece-
dented reduction in biodiversity. This history can serve to
place in perspective the nature of modern pertubations on
the Earth's environment and its biota. It is worthwhile to
contemplate that the current rate of extinction equals or ex-
ceeds that of the terminal Cretaceous event and constitutes
the greatest catastrophy ever imposed on the planet. As
Kenneth Hsu noted, "What should we be afraid of? Our
neighbors are not our enemy, nor or we likely to be exter-
minated by our nearest relatives. Our enemy is the disas-
trous disregard of the manmade catastrophe that threatens
to exterminate many species on earth, including, eventu-
ally, Homo sapiens" (1992). It is an exciting time for study-
ing, interpreting, and assessing the implications of vegeta-
tional, faunal, and environmental history. The broadest
scope of talent, methodologies, and context information
will be needed. A lot is riding on the outcome.
Origins of North American Biogeographic Affinities 333
References
Ager, D. 1993. The New Catastrophism. Cambridge Univer-
sity Press, Cambridge, U.K.
Arber, E. A. N. and J. Parkin. 1907. The origin of the an-
giosperms. Bot. J. Linn. Soc. 38: 29-80.
Arnold, C. A. 1947. An Introduction to Paleobotany.
McGraw-Hill, New York.
Avise, J. C. et al. (+ 7 authors). 1987. Intraspecific phyto-
geography: the mitochondrial DNA bridge between
population genetics and systematics. Annu. Rev. Ecol.
Syst. 18: 489-522.
Axelrod, D. I. 1975. Evolution and biogeography of the
Madrean-Tethyan sclerophyll vegetation. Ann. Mis-
souri Bot. Garden 62: 280-334.
Boufford, D. E. and S. A. Spongberg. 1983. Eastern
Asian-eastern North American phytogeographical re-
lationships—a history from the time of Linnaeus to
the twentieth century. Ann. Missouri Bot. Garden 70:
423-439.
Braun, E. L. 1955. The phytogeography of unglaciated east-
ern United States and its interpretation. Bot. Rev. 21:
297-375.
Breedlove, D. E. 1973. The phytogeography and vegetation
of Chiapas (Mexico). In: A. Graham (ed.), Vegetation
and Vegetational History of Northern Latin America.
Elsevier Science Publishers, Amsterdam, pp. 149-
165.
Bremer, K. 1992. Ancestral areas: a cladistic reinterpreta-
tion of the center of origin concept. Syst. Biol. 41: 436-
445.
Brooks, D. R. and D. A. McLennan. 1991. Phylogeny, Ecol-
ogy, and Behavior, a Research Program in Comparative
Biology. University of Chicago Press, Chicago.
Brown, C. A. 1938. The flora of Pleistocene deposits in the
western Florida Parishes, west Feliciana Parish, and
east Baton Rouge Parish, Louisiana. In: H. N. Fisk,
H. G. Richards, C. A. Brown, and W. C. Steere (eds.),
Contributions to the Pleistocene History of the Florida
Parishes of Louisiana. Department of Conservation,
Louisiana Geological Survey, New Orleans, LA. Geol.
Bull. 12, pp. 59-96.
Chaney, R. W. 1944. Pliocene floras of California and Ore-
gon. (Introduction and subsequent papers). Carnegie
Inst. Wash. Contrib. Paleontol. 553.
Coates, A. G. et al. (+ 7 authors). 1992. Closure of the Isth-
mus of Panama: the near-shore marine record of Costa
Rica and western Panama. Geol. Soc. Am. Bull. 104:
814-828.
Crawford, D. J. 1990. Plant Molecular Systematics, Macro-
molecular Approaches. Wiley-Interscience, New York.
Cronquist, A. 1987. A botanical critique of cladism. Bot.
Rev. 53: 1-52.
Davis, J. H., Jr. 1946. The peat deposits of Florida, their oc-
currence, development, and uses. Fla. Dept. Conserv.,
Geol. Surv., Geol. Bull. 30:1-247.
Davis, M. B. 1983. Quaternary history of deciduous forests
of eastern North America and Europe. Ann. Missouri
Bot. Garden 70: 550-563.
Deevey, E. S., Jr. 1949. Biogeography of the Pleistocene.
Bull. Geol. Soc. Am. 60: 1315-1416.
Delcourt, P. A. and H. R. Delcourt. 1977. The Tunica Hills,
Louisiana-Mississippi: late glacial locality for spruce
334 Late Cretaceous and Cenozoic History of North American Vegetation
and deciduous forest species. Quatern. Res. 7: 218-
237.
Donoghue, M. J. and E D. Cantino. 1988. Paraphyly, ances-
tors, and the goals of taxonomy: a botanical defense of
cladism. Bot. Rev. 54:107-128.
Fernald, M. L. 1925. Persistence of plants in unglaciated
areas of boreal America. Mem. Am. Acad. Arts Sci. 15:
241-342.
. 1931. Specific segregations and identities in some
floras of eastern North America and the Old World.
Rhodora 33: 25-63 [reprinted as Contrib. Gray Herb.
93].
Forey, P L. et al. (+ 5 authors). 1994. Cladistics, A Practical
Course in Systematics. Oxford Science Publications,
Claredon Press, Oxford, U.K. Systematics Association
Publ. 10.
Frederiksen, N. O. 1981. Middle Eocene to Early Oligocene
plant communities of the Gulf Coast. In: J. Gray, A. J.
Boucot, and W. B. Berry (eds.), Communities of the
Past. Hutchinson Ross Publishing Co., Stroudsburg,
PA. pp. 493-549.
Fritsch, P. 1996. Isozyme analysis of intercontinental dis-
juncts within Styrax (Styracaceae): implications for
the Madrean-Tethyan hypothesis. Am. J. Bot. 83:
342-355.
Gomez-Pompa, A. 1973. Ecology of the vegetation of Vera-
cruz. In: A. Graham (ed), Vegetation and Vegetational
History of Northern Latin America. Elsevier Science
Publishers, Amsterdam, pp. 73-148.
Graham, A. 1966. Plantae rariores Camschatcenses: a trans-
lation of the dissertation of Jonas P. Halenius, 1750.
Brittonia 18:131-139.
(ed.). 1972a. Floristics and Paleofloristics of Asia
and Eastern North America. Elsevier Science Publish-
ers, Amsterdam.
. 1972b. Outline of the origin and historical recogni-
tion of floristic affinities between Asia and eastern
North America. In: A. Graham (ed.), Floristics and Pa-
leofloristics of Asia and Eastern North America. Else-
vier Science Publishers, Amsterdam, pp. 1-18.
(ed.). 1973a. Vegetation and Vegetational History of
Northern Latin America. Elsevier Science Publishers,
Amsterdam.
. 1973b. History of the arborescent temperate ele-
ment in the Latin American biota. In: A. Graham (ed.),
Vegetation and Vegetational History of Northern Latin
America. Elsevier Science Publishers, Amsterdam, pp.
301-314.
. 1976. Studies in neotropical paleobotany. II. The
Miocene communities of Veracruz, Mexico. Ann. Mis-
souri Bot. Garden 63: 787-842.
. 1982. Diversification beyond the Amazon Basin. In:
G. T. Prance (ed.), Biological Diversification in the
Tropics. Columbia University Press, New York. pp.
78-90.
. 1985. Studies in neotropical paleobotany. IV. The
Eocene communities of Panama. Ann. Missouri Bot.
Garden 72: 504-534.
. 1987. Miocene communities and paleoenvironments
of southern Costa Rica. Am. J. Bot. 74:1501-1518.
. 1988a. Studies in neotropical paleobotany. V. The
lower Miocene communities of Panama—the Culebra
Formation. Ann. Missouri Bot. Garden 75:1440-1466.
. 1988b. Studies in neotropical paleobotany. VI. The
lower Miocene communities of Panama—the Cu-
caracha Formation. Ann. Missouri Bot. Garden 75:
1467-1479.
—. 1989. Studies in neotropical paleobotany. VII. The
lower Miocene communities of Panama—the La Boca
Formation. Ann. Missouri Bot. Garden 76: 50-66.
—. 1990. Late Tertiary microfossil flora from the Re-
public of Haiti. Am. J. Bot. 77: 911-926.
—. 1991a. Studies in neotropical paleobotany. VIII.
The Pliocene communities of Panama—introduction
and ferns, gymnosperms, angiosperms (monocots).
Ann. Missouri Bot. Garden 78: 190-200.
—. 1991b. Studies in neotropical paleobotany. IX. The
Pliocene communities of Panama—angiosperms (di-
cots). Ann. Missouri Bot. Garden 78: 201-223.
—. 1991c. Studies in neotropical paleobotany. X. The
Pliocene communities of Panama—composition, nu-
merical representations, and paleocommunity paleo-
environmental reconstructions. Ann. Missouri Bot.
Garden 78: 465-475.
—. 1992. Utilization of the isthmian land bridge dur-
ing the Cenozoic—paleobotanical evidence for timing,
and the selective influence of altitudes and climate.
Rev. Palaeobot. Palynol. 72: 119-128.
— and C. Heimsch. 1960. Pollen studies of some Texas
peat deposits. Ecology 41: 785-790.
Graham, A. and D. M. Jarzen. 1969. Studies in neotropical
paleobotany. I. The Oligocene communities of Puerto
Rico. Ann. Missouri Bot. Garden 56: 308-357.
Gray, A. 1840. Dr. Siebold, Flora Japonica. Am. J. Sci. Arts
39: 175-176.
. 1846. Analogy between the flora of Japan and that
of the United States. Am. J. Sci. Arts II, 2: 135-136.
-. 1859. Diagnostic characters of new species of
phanerogamous plants collected in Japan by Charles
Wright, botanist of the U.S. North Pacific Exploring Ex-
pedition. (Published by request of Captain James
Rodgers, Commander of the Expedition). With obser-
vations upon the relations of the Japanese flora to that
of North America, and of other parts of the northern
temperate zone. Mem. Am. Acad. Arts 6: 377—452.
Gray, J. 1960. Temperate pollen genera in the Eocene (Clai-
borne) flora, Alabama. Science 132: 808-810.
Halenius, J. P. 1750. Plantae Rariores Camschatcenses. The-
sis, University of Uppsala, Uppsala, Sweden.
Hedges, S. B., C. A. Hass, and L. R. Maxson. 1994. Reply:
Towards a biogeography of the Caribbean. Cladistics
10:43-55.
Hoey, M. T. and C. R. Parks. 1991. Isozyme divergence be-
tween eastern Asian, North American, and Turkish
species of Liquidambar (Hamamelidaceae). Am. J. Bot.
78: 938-947.
and . 1994. Genetic divergence in Liq-
uidambar styraciflua, L. formosana, and L. acalycina
(Hamamelidaceae). Syst. Bot. 19: 308-316.
Hou, H.-Yu. 1983. Vegetation of China with reference to its
geographical distribution. Ann. Missouri Bot. Garden
70: 509-548.
Hsu, K. J. 1992. Challenger at Sea, a Ship that Revolution-
ized Earth Science. Princeton University Press, Prince-
ton, NJ.
Hulten, E. 1927-1930. Flora of Kamtchatka and the Adja-
cent Islands, Parts I-IV Almquist & Wiksell, Stock-
holm.

. 1937. Outline of the History of Arctic and Boreal
Biota During the Quaternary Period. Thule, Stock-
holm.
Humphries, C. J. and J. A. Chappill. 1988. Systematics as
science: a response to Cronquist. Bot. Rev. 54: 129-
144.
and L. R. Parenti. 1986. Cladistic Biogeography. Ox-
ford Monographs on Biogeography. Clarendon Press,
Oxford, U.K. No. 2.
Hunter, R. L. and C. L. Markert. 1957. Histochemical
demonstration of enzymes separated by zone elec-
trophoresis in starch gels. Science 125:1294-1295.
Hunziker, J. H., R. A. Palacios, A. de Valesi, and L. Poggio.
1972. Species disjunctions in Larrea: evidence from
morphology, cytogenetics, phenolic compounds, and
seed albumins. Ann. Missouri Bot. Garden 59:
224-233.
Iversen, J. 1941. Land occupation in Denmark's Stone Age.
A pollen-analytical study of the influence of farmer
culture on the vegetational development. Danmarks
Geol. Unders. 66: 20-65.
Johnston, I. M. 1940. The floristic significance of shrubs
common to North and South American areas. J. Arnold
Arb. 21: 356-363.
Kalin Arroyo, M. T., R H. Zedler, and M. D. Fox (eds.). 1995.
Ecology and Biogeography of Mediterranean Ecosys-
tems in Chile, California, and Australia. Ecological
Studies 108. Springer-Verlag, Berlin.
Leopold, E. B. 1987. Past and future climatic and hydro-
logic patterns at Hanford, Washington. Northwest En-
viron. J. 3: 75-96.
Li, H.-L. 1952. Floristic relationships between eastern Asia
and eastern North America. Trans. Am. Philos. Soc.
42: 371-429.
Li, W.-H. and D. Graur. 1991. Fundamentals of Molecular
Evolution. Sinauer Associates, Sunderland, MA.
Manchester, S. R. 1995. Paleophytogeography of extant
eastern Asian "endemic" seed plant genera with paleo-
botanical records in the Tertiary of North America. In:
International Conference of Diversification and Evolu-
tion of Terrestrial Plants in Geological Time (ICTPG),
Nanjing, 1995. p. 82.
and S.-X. Guo. 1996. Palaeocarpinus (extinct Betu-
laceae) from northwestern China: new evidence for Pa-
leocene floristic continuity between Asia, North
America, and Europe. Int. J. Plant Sci. 157: 240-246.
Margoliash, E. 1963. Primary structure and evolution of cy-
tochrome C. Proc. Natl. Acad. Sci. USA 50: 672-679.
Marie-Victorin, E 1938. Phytogeographical problems in
eastern Canada. Am. Midland Nat. 19: 489-558.
Markert, C. L. and E M011er. 1959. Multiple forms of en-
zymes: tissue, ontogenetic and species-specific pat-
terns. Proc. Natl. Acad. Sci. USA 45: 753-763.
Martin, P. S. 1958. A biogeography of reptiles and amphib-
ians in the Gomez Farias region, Tamaulipas, Mexico.
Miscellaneous Publs. 101. Museum of Zoology Uni-
versity of Michigan, Ann Arbor, MI.
and B. E. Harrell. 1957. The Pleistocene history of
temperate biotas in Mexico and eastern United States.
Ecology 38: 468-480.
Marvin, U. B. 1973. Continental Drift, the Evolution of a
Concept. Smithsonian Institution Press, Washington,
D.C.
McVaugh, R. 1943. The vegetation of the granitic flat-rocks
Origins of North American Biogeographic Affinities 335
of southeastern United States. Ecol. Monogr. 13: 120-
166.
Meusel, H. Von. 1969. Beziehungen in der florendifferen-
zierung von Eurasien und Nordamerika. Flora 158:
537-564.
Miranda, E and A. J. Sharp. 1950. Characteristics of the
vegetation in certain temperate regions of eastern Mex-
ico. Ecology 31: 313-333.
Morrone, J. J. and J. V. Crisci. 1995. Historical biogeogra-
phy: introduction to methods. Annu. Rev. Ecol. Syst.
26: 373-401.
Nei, M. 1972. Genetic distance between populations. Am.
Nat. 106: 283-293.
. 1987. Molecular Evolutionary Genetics. Columbia
University Press, New York.
Nelson, G. 1978. From Candolle to Croizat: comments on
the history of biogeography. J. Hist. Biol. 11: 269-305.
and N. Platnick. 1981. Systematics and Biogeogra-
phy, Cladistics and Vicariance. Columbia University
Press, New York.
Nichols, N. 1997. Increased Australian wheat yield due to
recent climate trends. Nature 387:484-485.
Niemela, P. and W. J. Mattson. 1996. Invasion of North
American forests by European phytophagous insects.
BioScience 46: 741-753.
Orians, G. H. and O. T. Solbrig (eds.). 1977. Convergent
Evolution in Warm Deserts. Dowden, Hutchinson &
Ross, Stroudsburg, PA.
Page, R. D. M. 1988. Quantitative cladistic biogeography:
constructing and comparing area cladograms. Syst.
Zool. 37: 254-270.
and C. Lydeard. 1994. Towards a cladistic biogeog-
raphy of the Caribbean. Cladistics 10: 21-41.
Perry, J. P, Jr. 1991. The Pines of Mexico and Central Amer-
ica. Timber Press, Portland, OR.
Phipps, J. B. 1983. Biogeographic, taxonomic, and cladistic
relationships between east Asiatic and North Ameri-
can Crataegus. Ann. Missouri Bot. Garden 70: 667-
700.
Potzger, J. E. and B. C. Tharp. 1943. Pollen record of Cana-
dian spruce and fir from Texas bog. Science 98:
584-585.
. 1947. Pollen profile from a Texas bog. Ecology 28:
274-280.
. 1954. Pollen study of two bogs in Texas. Ecology
35:462-466.
Qiu, Y.-L., C. R. Parks, and M. W. Chase. 1995a. Molecular
divergence in the eastern Asia-eastern North America
disjunct Section Rytidospermum of Magnolia (Magno-
liaceae). Am. J. Bot. 82:1589-1598.
Qiu, Y.-L., M. W. Chase, and C. R. Parks. 1995b. A chloro-
plast DNA phylogenetic study of the eastern
Asia-eastern North America disjunct section Ryti-
dospermum of Magnolia (Magnoliaceae). Am. J. Bot.
82: 1582-1588.
Raven, P. H. 1963. Amphitropical relationships in the floras
of North and South America. Q. Rev. Biol. 38:
151-177.
. 1971. The relationships between "Mediterranean"
floras. In: P. H. Davis, P. C. Harper, and I. C. Hedge
(eds.), Plant Life of South-West Asia. Botanical Society
of Edinburgh, Edinburgh, Scotland, pp. 119-134.
. 1972. Plant species disjunctions: a summary. Ann.
Missouri Bot. Garden 59: 234-246.
336 Late Cretaceous and Cenozoic History of North American Vegetation
—. 1973. The evolution of Mediterranean floras. In: E
di Castri and H. A. Mooney (eds.), Mediterranean Type
Ecosystems, Origin and Structure. Ecological Studies.
Springer-Verlag, Berlin. Vol. 7. pp. 213-224.
— and D. I. Axelrod. 1975. History of the flora and
fauna of Latin America. Am. Sci. 63: 420-429.
-. 1978. Origin and relationships of the California
flora. Univ. Calif. Publ. Bot. 72: 1-134.
Riddle, B. R. and R. L. Honeycutt. 1990. Historical biogeog-
raphy in North American arid regions: an approach
using mitochondrial-DNA phylogeny in grasshopper
mice (genus Onychomys). Evolution 44:1-15.
Rieseberg, L. H. 1995. The role of hybridization in evolu-
tion: old wine in new skins. Am. J. Bot. 82: 944-953.
Ronquist, E 1997. Dispersal-vicariance analysis: a new ap-
proach to the quantification of historical biogeography.
Syst. Biol. 46: 195-203.
Rosenzweig, C. and D. Hillel. 1995. Potential impacts of cli-
mate change on agriculture and food supply. Conse-
quences, Nature Implications Environ. Change 1:
23-32.
Sarich, V. M. 1977. Rates, sample sizes, and the neutrality
hypothesis for electrophoresis in evolutionary studies.
Nature 265: 24-28.
Simpson, G. G. 1943. Mammals and the nature of conti-
nents. Am. J. Sci. 241: 1-31.
Simroth, H. 1914. Die Pendulations Theorie. 2nd edition.
K. Grethlein, Leipzig.
Solbrig, O. T. 1972. The floristic disjunctions between the
"Monte" in Argentina and the "Sonoran Desert" in
Mexico and the United States. Ann. Missouri Bot. Gar-
den 59: 218-223.
Styles, B. T. 1993. Genus Pinus: a Mexican purview. In: T. R
Ramamoorthy, R. Bye, A. Lot, and J. Fa (eds.), Biologi-
cal Diversity of Mexico, Origins and Distribution. Ox-
ford University Press, Oxford, U.K. pp. 397-420.
Sue, J. R 1984. Origin and evolution of the Mediterranean
vegetation and climate in Europe. Nature 307:429-432.
Thrower, N. J. W. and D. E. Bradbury (eds.). 1977.
Chile-California Mediterranean Scrub Atlas, A Com-
parative Analysis. Dowden, Hutchinson & Ross,
Stroudsburg, PA.
Tiffney, B. H. 1985. Perspectives on the origin of the floris-
tic similarity between eastern Asia and eastern North
America. J. Arnold Arb. 66: 73-94.
Udvardy, M. D. E 1969. Dynamic Zoogeography, with Spe-
cial Reference to Land Animals. Van Nostrand Rein-
hold, New York.
van der Hammen, T, T. A. Wijmstra, and W. H. Zagwijn.
1971. The floral record of the Late Cenozoic of Europe.
In: K. K. Turekian (ed.), The Late Cenozoic Glacial
Ages. Yale University Press, New Haven, CT. pp.
391-424.
Van Dyke, E. C. 1940. The origin and distribution of the
coleopterous insect fauna of North America. Proc.
Sixth Pacif. Sci. Congr. 4: 255-268.
Wagner, W. L. and V. A. Funk (eds.). 1995. Hawaiian Bio-
geography: Evolution on a Hot Spot Archipelago.
Smithsonian Institute Press, Washington, D.C.
Wehr, W. C. 1995. Early Tertiary flowers, fruits, and seeds of
Washington state and adjacent areas. Wash. Geol. 23:
3-16.
Wiley, E. O. 1988. Vicariance biogeography. Annu. Rev.
Ecol. Syst. 19: 513-542.
, D. Siegel-Causey, D. R. Brooks, and V. A. Funk.
1991. The Complete Cladist, A Primer of Phylogenetic
Procedures. University of Kansas Museum of Natural
History. Lawrence, KS, Special Publ. 19.
Willis, J. C. 1922. Age and Area: A Study in Geographical
Distribution and Origin of Species. Cambridge Univer-
sity Press, Cambridge, U.K.
. 1949. The birth and spread of plants. Boissiera 8:
1-561.
Wolfe, J. A. 1975. Some aspects of plant geography of the
northern hemisphere during the Late Cretaceous and
Tertiary. Ann. Missouri Bot. Garden 62: 264-279.
. 1978. A paleobotanical interpretation of Tertiary
climates in the northern hemisphere. Am. Sci. 66:
694-703.
-. 1979. Temperature Parameters of Humid to Mesic
Forests of Eastern Asia and Relation of Forests of Other
Regions of the Northern Hemisphere and Australasia.
U.S. Geology Survey, Reston, VA. Professional Paper
1106, pp. 1-37.
Wood, C. E., Jr. 1970. Some floristic relationships between
the southern Appalachians and western North Amer-
ica. In: P C. Holt (ed.), The Distributional History of
the Biota of the Southern Appalachians. Part II: Flora.
Virginia Polytechnic Institute State University (Blacks-
burg). Research Div. Monogr. 2, pp. 331-404.
. 1972. Morphology and phytogeography: the classi-
cal approach to the study of disjunctions. Ann. Mis-
souri Bot. Garden 59:107-124.
Wu, Z. 1983. On the significance of Pacific intercontinental
discontinuity. Ann. Missouri Bot. Garden 70: 577-
590.
Zuckerkandl, E. 1987. On the molecular evolutionary
clock. J. Mol. Evol. 26: 34-46.
and L. Pauling. 1962. Molecular disease, evolution
and genetic heterogeneity. In: M. Kash and B. Pullman
(eds.), Horizons in Biochemistry. Academic Press,
New York. pp. 189-225.
. 1965a. Molecules as documents of evolutionary
history. J. Theor. Biol. 8: 357-366.
. 1965b. Evolutionary divergence and convergence
in proteins. In: V. Bryson and H. J. Vogel (eds.), Evolv-
ing Genes and Proteins. Academic Press, New York,
pp. 97-166.
General Readings
Herrera, C. M. 1995. Plant-vertebrate seed dispersal sys-
tems in the Mediterranean: ecological, evolutionary,
and historical determinants. Annu. Rev. Ecol. Syst. 26:
705-727.
Nixon, K. C. 1996. Paleobotany in cladistics and cladistics
in paleobotany: enlightenment and uncertainty. Rev.
Palaeobot. Palynol. 90: 361-373.
Simpson, B. B. and J. Cracraft. 1995. Systematics: the sci-
ence of biodiversity. BioScience 45: 670-672.
Notes
1. A notable feature of these botanical dissertations
was, appropriately enough, the flowery language used in
the dedications. Halenius wrote:

To my foremost patron, the most Reverend and Re-
nowned Lord and Master Engelbert Halenius, Professor
of Sacred Theology, Distinguished member of the Eccle-
siastical and Academic Consistoires, Pastor in Danne-
marck.
Your exceptional favor and kindness, showered
abundantly and continually on me, have induced me,
my distinguished patron, to enhance my modest labors
with your renowned name. For as soon as it was my
good fortune, after the loss of my former director of
studies, to come to this temple of Minerva, you saw fit,
noble sir, to receive me most graciously, undeserving
though I was. You have been willing not only to look out
for my fortunes in every way, but you deigned also to
further all my undertakings. In view of these evident to-
kens of your favor, I shall always be deeply devoted to
you. I therefore now offer you the first-fruits of my stud-
ies, and I earnestly urge you to accept them graciously
and in the future also bestow upon me your continued
favor. As long as I live I shall never cease to offer to
Almighty God the most fervent prayers that happiness
may come to you and yours. Your Reverence's most
humble servant. (1750)
Particularly poignant for those who have provided a
modern-day education to our own issue is the dedication
to the
[M]ost Reverend and Learned Peter Halenius, most vig-
ilant Co-Minister of the Church of Hedesunda, and my
most indulgent father, with all due reverence until the
day of my death.
The day has finally dawned [!] when I can publicly
bring to an end my academic apprenticeship. The work
which is, as it were, the first-fruits of that period of
study—a work which I owe to your eager interest and
financial assistance [!!]—I unhesitatingly dedicate to
you, since in a sense it makes that dedication of its own
accord. The countless benefits you have shown me from
my earliest years are too great to be recompensed by so
Origins of North American Biogeographic Affinities 337
modest a work. But since I am well aware, my dear fa-
ther, that you ask nothing more from your son than a
grateful heart, I earnestly ask you to accept now with fa-
therly affection these pages which bear witness to my
sentiments of respect for you. For the rest, be sure, fa-
ther, that more filial piety for you will ever be deep in
my heart than I can ever express in words. With devoted
prayers to Almighty God for your unimpaired health
and long life, I shall remain until death, Your most obe-
dient son. (1750)
I mentioned to a class that such expressions of rever-
ence for their mentors in theses has regrettably passed out
of fashion. A student in that class later did graduate work
with me, and when her thesis was submitted it contained
the following dedication:
Discipula deterrima, magna cum modestia, summissa
animo, animo grata omniscienti, benignissimo, genero-
sissimo consiliario, cui libellum dedicatum est, gratiam
referre vult. Discipula nequissima quae ad pedes prae-
clari magistri prosternata est, colens praesentiam eius et
laudans sapientiam, elegantiam, urbanitatem, humani-
tatemque immensam, magnum gaudium, decus mag-
num verbis apte exprimere non potest. Multitude cae-
lestis ilium propriosque defendant.
It is with great humility that this most worthless stu-
dent wishes to convey a cornucopia of gratitude to the
most gracious, all knowledgeable, most munificent of
mentors, to whom this humble essay is dedicated. Mere
words cannot adequately express the great joy and
honor of this lowly servant, who prostrates herself at the
feet of said illustrious master in homage to his presence
and in praise of his infinite wisdom, gentility and re-
finement. May the celestial hosts glorify his name and
smile upon his ancestors, as surely he is descended
from the heavens.
This is a remarkably perceptive piece of work.

Organisms are indexed to genus with fossil occurrences indicated separately. See Contents for major subject headings.
Extinct plant formations are indexed; for extant plant formations see vegetation summaries (chapters 5-8) and Epilogue.
Notes and Prologue are not indexed.
Abies, 8, 9,10,11,12,13,14,15, 55,
121, 159, 195, 205, 266, 282
fossil, 121, 204, 210, 211, 214, 215,
220, 221, 222, 226, 227, 228, 229,
233, 249, 251, 252, 254, 255, 256,
259, 260, 261, 262, 263, 264, 265,
266, 268, 281, 284, 285, 286, 287,
289, 290, 292, 293, 295, 296, 297,
298, 299, 300, 302, 303, 326, 327
Acacia, 19, 24, 25, 295
fossil, 295
Acer, 13,14,15,16, 17, 18, 21, 120,
123, 149, 150, 157, 282, 285, 306,
322, 324, 325, 326
fossil, 165, 168, 173, 174, 195, 196,
199, 202, 203, 206, 210, 211, 214,
215, 217, 220, 221, 222, 226, 228,
229, 231, 249, 250, 251, 252, 255,
256, 258, 259, 260, 261, 262, 263,
266, 281, 286, 288, 289, 291
acid rain, 10
Acropora, 38
Acrostichum, fossil, 194, 195, 196, 200,
202, 266
Actinidia, fossil, 173, 213, 214, 227
active margins, 94
Adenostoma, 21
fossil, 296, 297
Aesculus, 16, 22, 285, 306, 319, 322
fossil, 123, 204, 210, 214, 229, 255
Agave, 25, 320
fossil, 294
age and area hypothesis, 318
age-area curve, 223
Aglaoreidia, fossil, 196, 219
Agnotocastor, fossil, 191
Agropyron, 19, 20, 23
Agulhas current (See also ocean cur-
rents), 34-35
Ailanthipites, fossil, 196
Ailanthus, fossil, 200, 202, 206, 208,
210,254,259,262,323
INDEX
Alangium, fossil (See also
Lanagiopollis), 107,162,173,174,
194, 196, 213, 214, 223, 225, 226,
249, 250, 323, 324
Alaska Range, 52, 57, 242, 262
albedo, 94, 144-145, 189, 278, 279
Albertipollenites, fossil, 196
Alchornea, fossil, 200
Alectoria, 7
Aleurites, fossil, 173, 202
Aleutian arc, islands, trench, 52, 57,
61
Alfaroa, fossil, 163,164,195,198,
249
alkenones, 32, 88, 332
Allantoidopsis, fossil, 167
Alleenbya, fossil, 216
Alligator, fossil, extant, 98,106
allochthonous material, 120
Allophylus, fossil, 202, 214
allozyme(s), 328
Alnipollenites, fossil, 164-165
Alnobetula, fossil, 227
Alnus, 7, 10, 11, 14, 15, 16, 123, 255,
285,326
fossil, 107, 121, 163, 164, 167, 168,
169, 173, 174, 177, 194, 195, 196,
198, 199, 200, 204, 210, 214, 215,
217, 221, 222, 223, 225, 226, 227,
228, 233, 249, 250, 251, 252, 254,
255, 258, 259, 260, 261, 262, 263,
264, 265, 266, 281, 283, 286, 289,
290, 291, 302, 303, 304, 305, 323,
327
Alopecurus, 7
Alternaria, fossil, 259
altithermal. See postglacial thermal op-
timum
Amaranthus, fossil, 261
Ambrosia, 24, 295
fossil, 249, 250, 251, 256, 259, 283,
287, 288, 289, 293, 299
338
Amelanchier, 21
fossil, 170, 204, 206, 210, 228, 229,
255,259,262
Amelanchites, fossil, 167
Amentotaxus, fossil, 215, 323
Amesoneuron, fossil, 167
Amoenitates Academicae, 322
Amorpha, fossil, 228, 253, 254, 255, 293
Ampelocissus, fossil, 213, 214
Ampelopsis, fossil, 213, 216
Amphananthe, fossil, 213
Amphelopsis, fossil, 167,169
Amphictis, fossil, 191
Anacardites, fossil, 202
Anacolosa, fossil, 196
Anamirta, fossil, 174, 213
Anchitheriomys, fossil, 102,191
Anchitherium, fossil, 100
ancient DNA, 255,332
Andromeda, fossil, 155, 175, 227
Anemia, fossil, 194, 195, 196
Annona, fossil, 214, 221
Anoda, fossil, 259
Anomone, 322
Anonaspermum, fossil, 213
Antarctica
early glaciation, 86, 87, 89, 187, 188,
332
full glaciation, 190,198
separation, 187
sudden warming, 198
Antennaria, fossil, 291
Apeiba, fossil, 200
aphelion, 29
Apocynophyllum, fossil, 167
Apocynospermum, fossil, 202
Appalachian Mountains, 45, 142, 187,
189,242
Aquilapollenites, fossil, 150,155,156,
158
Aralia, 285
fossil, 150, 170, 194, 227

Araliophyllum, fossil, 202
Arbutus, 14, 15, 22, 319, 320
fossil, 123, 220, 228, 229, 253, 259,
262
Archaeoleersia, fossil, 250
Archeampelos, fossil, 170
Arctagrostis, 7
Arctic glaciations, 89,188, 242, 245, 262
Arctophila, 7
Arctostaphylos, 7, 13, 21
fossil, 107, 204, 210, 227, 250, 253,
254, 262, 296, 297
area cladograms, 109, 329-330, 331
Arecipites, fossil, 196
Aristida, 17,19
Aristolochia, fossil, 202, 206
Artemisia, 13, 20, 21, 22, 23, 24, 295
fossil, 226, 249, 250, 251, 252, 259,
260, 265, 268, 289, 290, 291, 292,
293, 295, 296, 297, 298, 299, 300,
302, 303, 304, 305, 324, 331
Artocarpium, fossil, 107
Artocarpus, fossil, 150
Asimina, 18, 285, 323
Asplenium, 322
fossil, 202
association, plant
definition, 7
temporal nature. See community
stability
Aster, fossil, 249
Asterocarpinus, fossil, 206, 221, 222
asteroid
Cretaceous impact, 65-68, 88, 146,
159-161, 175
definition, 85
Eocene impact, 189, 190
Miocene impact, 223
Astronium, fossil, 202, 206, 214
Atacama Desert, 33
Atamisquea, 320
Athyana, fossil, 202, 206
Atitlan spike, 277
atmospheric pressure, 30
Atriaecarpum, fossil, 213
Atriplex, 23, 24
fossil, 134, 298
Aulacomnium, 7
Aulacoseira, fossil, 261
aurora australis, borealis, 35
autochthonous material, 120
Avena, 20
Averrhoites, fossil, 157, 165,167,169,
170
Avicennia, 25, 195, 299
Azolla, fossil, 167, 168, 170, 202
Baccharis, 306,
fossil, 253, 293
Bahia, fossil, 135
bajadas, 24
Barbados corals (See also Acropora),
38-39, 86, 88, 279, 283
Barghoornia, fossil, 215
Bamebyanthus, fossil, 171
Barringtonia, fossil, 174
Basin and Range Province, 50
Basopollis, fossil, 150
Bauhinia, 117, 119
Beilschmiedia, fossil, 202
Benioff zone(s), 44, 57
Berberis, 14, 21
fossil, 220, 221, 255
Berchemiopsis, fossil, 202
Bergmann's rule, 99,133
Bering Straits, Beringia, land bridge,
60-61, 101, 175, 190, 191, 242,
320,330
Berriochloa, fossil, 250
Bersericarpum, fossil, 213
Betula, 7, 10, 11, 16, 120, 282, 285, 306
fossil, 121, 126, 163, 164, 165, 168,
169, 174, 177, 194, 198, 199, 200,
204, 210, 214, 215, 217, 220, 222,
223, 226, 227, 228, 229, 233, 246,
249, 250, 251, 252, 254, 255, 258,
259, 260, 262, 263, 264, 265, 266,
281, 286, 287, 288, 289, 290, 291,
298, 302, 303, 304, 305, 323
biological carbon cycle, 43
Biorbia, fossil, 250
Bistorta, 12
Blackia, fossil, 102,191
blackland prairie, 19
Boehlensipollis, fossil, 251
Bohlenia, fossil, 215
Boisduvalia, 327
boreotropics concept, flora, 109, 175,
319,332
Botrychium, fossil, 168, 226, 251
Botryococcus, fossil, 255, 259, 280
Bougainvillea, 320
Bouteloua, 19, 20, 295
Brachypotherium, fossil, 104
Bromus, 20, 23
Brooks Range, 57, 245
Brunhes-Matuyama, chron, magetic re-
versal, 98, 245
Buchloe, 19, 295
Buckleya, 322
Bulnesia, 320
Bumelia, 285
fossil, 150, 196, 213, 249
Bursera, 122, 283
fossil, 202, 206, 210, 211
Buxus, 319, 320
Caesalpinia, 194, 200
Caesalpinites, fossil, 202, 206
Calamagrostis, 12
Caldesia, fossil, 223
Callierogon, 7
Index 339
Callitris, 157
fossil, 175
Calocedrus, 13,14, 15
fossil, 255
Caltha, 12
Calycites, fossil, 215
Calycocarpum, fossil, 213, 215
Cananga, fossil, 174
Canavalia, fossil, 200
Cannabis, 120
Canticocculus, fossil, 165
Caprifoliipites, fossil, 196
carbon-3/4 plants, 42-43, 250
carbon cycle. See geochemical carbon
cycle
carbon dioxide
as amplifier of Milankovitch cycles,
39,44
effect of plate tectonics, 44
effect on climate, 41-44
in ancient air bubbles, 43, 278, 279
in Cretaceous
in foraminifera, 43
in middle Eocene-early Miocene,
187
in soil nodules, 42
Quaternary, 278
re stomatal analysis, 128
carbon isotopes, ratios
in pack rat middens, 133,137
in tooth enamel, 88, 167, 198, 250
in tree rings, 129
Monterey Formation, 254
Carcharias, fossil, 98
Cardiospermum, fossil, 200, 202, 206
Carex, 6, 7,12
fossil, 227, 250, 259, 263
Caricoidea, fossil, 223
Carnegiea, 24
fossil, 295
Carpinus, 285, 324, 326
fossil, 121, 194, 199, 215, 222, 226,
228, 229, 249, 250, 259, 261, 263,
283,284,290
Carya, 17, 18, 285, 320, 323, 326
fossil, 121, 123, 157, 162, 163, 164,
167, 169, 174, 177, 194, 195, 196,
197, 198, 199, 200, 204, 206, 215,
217, 223, 225, 226, 228, 229, 246,
249, 250, 251, 252, 255, 256, 259,
260, 261, 262, 263, 264, 281, 283,
284, 286, 287, 288, 289, 290, 291,
323
Caryapollenites, fossil, 164,168
Cascade Mountains, 52, 57,142, 204,
214,222,242, 245,258,301
Cassia, fossil, 123, 196, 210
Cassiope, 7
Castanea, 16, 288
fossil, 164, 194, 199, 204, 206, 208,
214, 217, 222, 223, 225, 226, 229,
340 Index
Castanea, fossil (continued)
233, 249, 251, 255, 259, 260, 263,
285,286, 287, 289,291,323
Castaneoidea, fossil, 194
Castaneophyllum, fossil, 215
Castanopsis, 13, 14, 15
fossil, 213, 254, 259, 323
Castor, fossil, 102
Catalpa, fossil, 221
Cathedra, fossil, 196
Ceanothus, 13, 21
fossil, 253, 254, 260, 295, 296, 297
Cedrela, fossil, 150, 173, 202, 206, 210,
217, 222, 228, 229, 251, 254, 255,
256,258, 259, 260, 262
Cedrelospermum, fossil, 206, 213, 215
Cedrus, fossil, 168, 196, 199, 255, 259,
261
Celastrus, 322
fossil, 107, 123, 125, 202, 206, 250
Celtis, 285, 320
fossil, 152, 164, 194, 195, 196, 200,
202, 204, 206, 213, 214, 218, 229,
233, 249, 250, 251, 255, 259, 266,
284,286,323,327
Cenchus, 122
Cephalanathus, fossil, 283
Cephalogale, fossil, 191
Cephalotaxus, fossil, 163, 210, 215, 259,
324
Ceratiola, fossil, 283
Ceratoides, fossil, 135
Ceratophyllum, fossil, 164,194, 200,
250
Cercidiphyllum, fossil, 163, 164, 165,
167, 168, 169, 174, 175, 199, 200,
204, 210, 214, 215, 221, 222, 255,
262, 263, 323,324
Cercidium, 24, 320
fossil, 253, 295
Cercis, 285, 319, 320, 326
fossil, 206, 220, 221, 222, 249
Cercocarpus, 7, 13, 20, 21, 22
fossil, 206, 210, 219, 220, 221, 228,
229, 253, 254, 262, 296
Cervus, fossil, 104
Cetraria, 7
Chaetoptelea, fossil, 165, 168, 200
Challenger (research vessel), 37-38
Chamaebatia, 13
fossil, 228
Chamaebatiaria, fossil, 220
Chamaecyparis, 7, 10, 14, 18, 323
fossil, 174, 200, 204, 205, 206, 210,
211, 214, 215, 228, 233, 251, 255,
262, 285,305, 327
Chamaedaphne, fossil, 227
Chamaesaracha, fossil, 135
Chandlera, fossil, 213
chaos, 32, 52, 53
Chara, fossil, 226
Charadrius, 321
charcoal, 281, 298
Chenopodium, fossil, 227, 250, 287
chestnut blight. See Endothia
Chicxulub Crater, 68, 160
Chilopsis, fossil, 220, 253
Chionanthus, 285
Chiranthodendron, fossil, 211
Chondrophyllum, fossil, 155,175
chrons, 98
Chrysolepis, fossil, 228, 296
Chrysophyllum, fossil, 152, 196
Chrysothamnus, 23, 295
fossil, 295
Cicatricosisporites, fossil, 156,196
Cinnamomophyllum, fossil, 107
Cinnamomum, fossil, 107,152, 164,
165, 168, 173, 225
circum-Antarctic current, 187
Cissites, fossil, 157
cladistics, 329, 331, 332
Cladonia, 7,10
Cladrastis, fossil, 222, 251, 252
CLAMP, 127, 128, 160, 161, 174, 207,
208, 210, 212, 214, 217, 219, 222,
228, 252,332
Clarnoxylon, fossil, 214
Cleome, fossil, 227
Clerodendrum, fossil, 174
Clethra, 17, 18, 195
Cleyera, fossil, 213, 223, 225
Cliftonia, fossil, 194, 196, 197, 251
CLIMAP, 86, 88, 155
climax vegetation, 306, 332
cloud cover, 28, 145
Clovis site, 295
Cnemidaria, fossil, 167,170
coastline (See also epicontinental sea)
Eocene, 148
Pliocene, 245
Coast Ranges, 142, 214, 242, 245, 262
Cocculus, fossil, 165, 251, 262
COHMAP, 295
Colubrina, fossil, 206
Comicilabium, fossil, 213
community stability (See also modern
analog method, geoflora concept),
15-16, 107-109, 195,280,
288-290, 292, 299, 304, 305, 306,
328, 331, 332
Comptonia, 17
fossil, 204, 215, 227, 251
Condalia, 24, 320
fossil, 220, 253
Conocarpus, 25, 229
continental configurations
Cretaceous, 145
Eocene, 147
Paleocene, 146
continentality, 93-94, 189
conveyor belt. See thermohaline cell
Convolvulites, fossil, 206
Conzattia, fossil, 206
Cordia, fossil, 251
Cordilleran ice sheet, 275, 276, 278,
292, 299, 301
Coriolis effect, 30
Cornophyllum, fossil, 167, 170
Cornus, 13, 18, 285, 322, 326
fossil, 107, 164, 165, 167, 204, 210,
213, 214, 215, 249, 250, 255, 259,
263, 286,323
Corylites, fossil, 323
Coryloides, fossil, 213
Corylopsis, fossil, 215
Corylus, 123, 285
fossil, 163, 164, 165, 166, 168, 174,
194, 199, 210, 215, 223, 225, 226,
229, 249, 251, 255, 259, 263, 264,
286, 291,323, 324
Costea, fossil, 197
Cotinus, fossil, 206
Coville foredeep, 57
Cowania, 21
Craigia, 214
fossil, 215, 221, 323
Cranwellia, fossil, 156
Crataegus, 285, 330
fossil, 206, 210, 215, 220, 221, 222,
226,228, 251,255, 259,293
Crepetocarpon, fossil, 194
Cressa, 320
Cruciptera, fossil, 163, 199, 213, 214,
215,222
Crudia, fossil, 194
Cryptantha, 320
fossil, 250
Cryptomeria, fossil, 196
Cucurbita, fossil, 287
Cunninghamia, fossil, 323
Cupania, fossil, 196, 214
Cupanieidites, fossil, 196
Cuphea, 117,118
Cupressacites, fossil, 219
Cupressus, 319
fossil, 253, 259
Cupuliferoidaepollenites, fossil, 196
Cupuliferoipollenites, fossil, 219
Curvitinospora, fossil, 213
Cyathea, fossil, 249
Cyclocarya, fossil, 165, 323
cyclones, 30
Cyclotella, fossil, 261
Cynelos, fossil, 191
Cyperacites, fossil, 205
Cyperites, fossil, 259
Cyperocarpus, fossil, 250
Cyperus, fossil, 263
Cyrilla, 18
fossil, 194, 196, 197, 223, 249, 251,
324
Cyrillaceaepollenites, fossil, 196
 Index 341

Dactylina, 7 Ditrichum, 7 Eorhiza, fossil, 214, 216

Dalbergia, fossil, 152, 167
Dalea, 24
Dansgaaxd-Oeschger (D-O) events, 40,
274,277
Daphne, fossil, 206
Dasylirion, 320
fossil, 294
Dasypus, fossil, 104, 106
Datisca, 319, 320
Davalia, fossil, 259
Davisicarpum, fossil, 213
Deccan Traps, 60, 63, 66, 142
deciduousness, 146, 161-162
Decodon, fossil, 213, 214, 216, 226
deforestation, effects on climate, 70-71
degassing, 44
DeGeer route, 63, 101, 147, 172
dendrochronology, 115, 128-131, 332
Cenomanian, 157
Eocene, 214
hydrogen isotopes, 295
Maastrichtian, 152, 155, 156, 158
Mio-Pliocene, 263
Paleocene, 163
Quaternary, 293, 296
Dendropanax, fossil, 170, 194, 200
Dennastra, fossil, 170
Dennstaedtia, fossil, 163, 170
Dennstaedtiopsis, fossil, 216
Dermatophyllites, fossil, 155, 175
Deschampsis, 12
Desmanthus, 320
deuterium, in dendrochronology, 129
Deviacer, fossil, 213, 215
Devils Hole, 39
diagenesis, 87
diatomite(s), 33-34, 100, 101, 257
Dichrostachys, 109
Dicotylophyllum, fossil, 167, 170
Dicmnum, 7
Diegansage, 253, 296
Diervilla, fossil, 63, 174, 199, 227
differential production, of pollen and
spores, 120
Dimorphandra, fossil, 171
Diospyros, 17, 18, 110, 177, 285, 322,
323
fossil, 155, 175, 196, 249, 251, 254,
255,259, 262
Diplazium, fossil, 216
Diploclisia, fossil, 213
Diplodipelta, fossil, 206
Diplopanax, fossil, 323
Diploporus, fossil, 213
Diplotropis, fossil, 194
Dipodomys, 102
Diporoconia, fossil, 211
Dipteronia, fossil, 173, 202, 323
Distylium, fossil, 202
Ditaxocladus, fossil, 323
divergence time, 320 eb'tvos, 319
DNA, sequencing, 320 epeirogeny, 142
Dodonaea, fossil, 206, 253 Ephedra, 23, 24, 295
Dombeya, fossil, 165, 167 fossil, 121, 194, 195, 196, 197, 200,
Draba, 4 204, 205, 210, 233, 246, 249, 252,
Drepanocladus, 7 253,259, 286,293
drilling mud thinners, 121 Ephedripites, fossil, 251
Drimys, fossil, 214 epicontinental sea, 33,144
Dromomeryx, fossil, 100 Cretaceous, 154, 175, 331
Drosera, fossil, 265 end of early Miocene, 188-189
Drumhellera, fossil, 158, 175 middle Eocene, 148
Dryas, 7 epidemic disease, 288, 306, 328
fossil, 291 Epilobium, 327
drying trend, 45, 187, 189 fossil, 259, 263
Dryophyllum, fossil, 171,194,195,196 Epipremnum, fossil, 227
Dryopteris, fossil, 107, 205 equiformal progressive areas, 318
Duckeophyllum, fossil, 194 equinox, 29, 39
Dulichium, fossil, 63 Equisetum, 7
Dupontia, 7 fossil, 156, 158, 163, 165, 167, 168,
dust, effect on climate, 39, 40, 59, 142 170, 175, 202, 205, 249, 259
Eocene, 170 Eremotherium, fossil, 280
Neogene, 242 Erethizon, fossil, 104
Quaternary, 278, 279 Erigeron, 12
dust bowl, 19 Eriodictyon, 21
Eriogonum, 21
easterlies (winds), 30 fossil, 296
East Pacific Rise, 47,51 Eriophorum, 6, 7
Echinocactus, 25 erosion, CO2 concentration. See silicate
Elatocladus, fossil, 163, 170 rocks
El Chichon, 57, 58, 59 Errazurizia, fossil, 320
Eleiosina, fossil, 221 Erythrina, fossil, 102
Eleocharis, 7 Ethela, fossil, 214, 216
Eleofimbris, fossil, 250 Eucnide, 320
Eleopoldia, fossil, 221 Eucommia, fossil, 172, 194, 200, 204,
Eliasofructus, fossil, 194 206, 210, 214,215,221
El Nino, 30-32 Eugenia, fossil, 125, 152, 167, 195, 197,
El Nino-Southern Oscillation (ENSO), 202, 206, 228, 262
30,31,59 Euodia, fossil, 222
Elomeryx, fossil, 191 Euonymus, 285
Elymus, 19, 20, 23 fossil, 210, 250
Emiliania, 32, 88 Euphorbia, fossil, 206
Emmenopterys, fossil, 212, 213, 215 Euphorbiophyllum, fossil, 152, 154
Empetrum, 7 Euptelea, fossil, 173
fossil, 291 Eurotia, 23
Emplectocladus, fossil, 253 Eustigma, fossil, 173
Encelia, 24, 295 Evodia, fossil, 259
fossil, 295 Evolvulus. 320
Endothia, 16 exclusion experiments, effects on vege-
Engelhardia, fossil, 121, 163, 164, 174, tation, 102
197, 198, 199, 200, 202, 203, 217, Eysenhardtia, fossil, 253
219, 221, 222, 223, 225, 249, 323
Ensete, fossil, 109, 213, 214, 215 Fagonia, 320
enthalpy, 209, 252 Fagopsis, fossil, 157, 167, 169, 175, 206,
Eoceltis, fossil, 194 207,215
Eocene termination date, 189-190 Fagus, 16, 17, 120, 128, 282, 285, 325,
Eoenglelhardia, fossil, 194 326
Eohypserpa, fossil, 213 fossil, 152, 194, 195, 198, 199, 210,
Eokachyra, fossil, 194 214, 217, 222, 225, 226, 229, 246,
Eomimosoidea, fossil, 194, 218 249, 250, 251, 254, 255, 258, 259,
Eomys, fossil, 191 262, 263, 283, 284, 288, 289, 291
342 Index

Fallugia, 21 Buffalo Canyon, 192, 194, 204, 212, Golden Valley, 104, 144, 150, 163,
fossil, 220 226,229, 232,233, 252 165, 166, 168
Farallon plate, 47, 50, 51, 57, 60, 165 Bull Run, 192, 211, 212, 232 Bear Den, 144, 150, 163, 165,
faunas, fossil Canyon, 144 166, 168, 176
Ash Beds, 250 Carbon Mt., 217 Camels Butte, 144, 150, 163, 165,
Badlands, 191 Carmel, 64, 245, 248, 254 166, 168, 176
Big Bend, 191 Catahoula, 192, 217, 218, 232 Gorge, 245
Bighorn Basin, 198 Chalk Bluffs, 144, 150, 152, 174, 192, Goshen, 192, 209, 212, 214, 232
Big Springs, 250 200, 212, 214 Green River, 101-102, 104, 125, 127,
Bug Creek, 145, 146, 156, 160 Chain's Volcanics, 192, 204 174, 192, 194, 200, 201, 203, 204,
Clovis, 285 Charlotte Ridge, 217 207, 209, 210, 211, 217, 219, 230,
Edwards Plateau, 285, 294 Chickaloon, 152,168, 217 231, 233,253, 254
Ellesmere Island, 98, 147, 172, 198 Chloropagus, 245, 248, 253 Gubik, 245, 248, 264, 265
extinctions, 104, 280 Chuckanut, 144, 150, 167 Haughton, 194, 223, 226, 232
Great Plains, 98 Chula Vista, 245, 246, 248, 254, 265, Hell Creek, 144, 152, 153, 154, 156,
Green River, 200 266, 267 160, 161, 168, 186
Hell creek, 101,145,156 Circle, 263 Holmatindur, 245, 248, 249, 267
Lance, 98,145,156 Claiborne, 104, 192, 194, 195, 197, Horseshoe Canyon, 144,152,153,
Laredo, 98 200, 218,229,231,266 158,168
migrations into North America, 105 Clarkia, 223, 245, 246, 248, 255, 256, House Ranch, 209
Rancho La Brea, 296 265 Idaho Group, 245, 266, 267
Scollard, 145 Clark's Fork, 144, 150, 166, 168 Jackson, 192,194, 195, 198, 229, 231,
Willwood, 166 Clarno, 109, 123, 132, 192, 194, 212, 266,303
Wind River, 174 213,214, 221,231 Joffre Bridge, 144, 168, 176, 212
FAUNMAP, 106 Coalspur/Scollard, 144, 150, 152, John Day, 194,221,222
fellfields, 7, 8 153, 154, 156, 160, 168 Kap K0benhavn, 245, 248, 249, 267
Fendlera, fossil, 220 Colgate, 152, 153 Katalla, 217
fern spike, 160 Comstock, 192, 212, 214, 232 Kilgore, 126, 245, 246, 248, 251, 252,
Festuca, 20, 23 Cook Inlet, 245, 248, 262, 267 266, 267, 324
Ficus, 152, 283 Cooper Pit, 152, 154 Kirkwood, 245, 248, 267
fossil, 157, 164, 168, 195, 214, 253 Copper Basin, 173, 192, 210, 211, Kissinger Lakes, 192, 200, 207, 209,
fire 212,214, 232 231
coniferous forest, 8 Creede, 192, 219, 223, 232 Kivalina, 245, 248, 264, 266, 267
in Aftonian period of aridity, 306 Dawson, 152 Klondike Mountain, 194, 214
savanna, 22 Denver, lower Denver, 144,150, 152, Kulthieth, 217
Firmiana, fossil, 173 153, 154, 155, 160 Kupreanof Island, 174
flavonoids, fossil, 257, 332 Deschutes, 245, 246, 248, 261, 265 Kushtaka, 152, 174, 176
floral provinces, 144, 147, 150, 162 Eagle Creek, 230 Lake Hazen, 192
stages, 212, 262, 264 Eastgate, 204, 229, 252 Lance, 144, 150, 152, 153, 154
floras, fossil, Cretaceous, Tertiary Elko, 192 La Porte, 192, 212,232
Aldrich Station, 245, 248, 253, 265 Ellensburg, 245, 246, 248, 258, 260, Laramie, 144, 150, 152, 153, 154
Alvord Creek, 192, 194, 212, 229, 233 262, 265, 267 Laredo, 192
Anaverde, 245, 248, 253 Ellesmere Island, 159, 192, 198, 199, Latah, 230, 245, 246, 248, 260, 262
Aniakchak, 217 231 Lava Camp, 245, 248, 263, 264, 265,
Arkose, 152 Elsinore, 192, 211, 231 266, 267
Ash Hollow, 245, 246, 248, 250, 267 Eureka Sound, 158, 168, 174, 177, Ledger, 245, 248, 266
Atanikerdluk, 144, 152, 153, 154, 155, 198 Legler, 249, 267
163,324 Fallen, 245, 248, 253 Little Mountain, 192, 200, 231
Beaufort, 104, 226, 249 Faraday, 230 Littleton, 154
Beaverhead Basins, 192, 194, 200, Fingerrock, 245, 246, 248, 262, 267 London Clay, 172
210, 223 Florissant, 56,102,104,120,121, Lower Cedarville, 192, 212, 232
Berg Lake, 107 125, 127, 192, 194, 204, 205, 207, Lower Edmonton, St. Mary River,
Bonnet Plume, 168 208, 209, 210, 223 152, 156, 168
Brandon, 104, 109, 192, 194, 223, 49-Camp, 245, 246, 248, 262 Marsland, 226
226, 229, 232, 249, 266 Fossil, 192, 221, 222, 223, 232 Martha's Vinyard, 245, 246, 248, 266,
Brandywine, 245, 246, 248, 249, 250, Ft. Union, 144, 150, 152, 160, 163, 267
266,267 164, 165, 166, 167, 168, 176, 323 Mary Sachs Gravel, 194, 226, 232, 249
Bridge Creek, 192, 212, 214, 221, 222, Furnace Creek, 245, 248, 253 Mascall, 245, 246, 248, 255, 262, 265
223,230,232 Genesee, 152 Mays Landing, 188, 192, 198, 246
Bryn Mawr, 249 Germer, 192, 204, 210, 231 McNairy Sand, 144, 150, 152, 153,
Buchanan Lake, 174,199 Gillam Springs, 245, 248, 262 154, 160, 186
 Index 343

Medicine Bow, 152, 154 Teewinot, 245, 246, 248, 252, 265, 267 French Lakes, 291
Metzel Ranch, 192, 210, 223 Tehachapi, 192,194, 212, 229, 233, 253 Cause Bog, 281
Middlegate, 204, 212, 229, 252, 320 Temblor, 245, 248, 254 Glacier National Park, 298
Middle Ravenscrag, 152 Thunder Mountain, 150, 173, 192, Goshen Springs, 287
Mint Canyon, 64, 331 200,204, 210, 232 Grand Teton National Park, 298
Modelo, 245, 248, 254 Thyra 0, 144, 152, 163, 175, 324 Hager Pond, 300
Morman Creek, 192, 210, 223 Tipperary, 192, 200, 231 Heuco Mountains, 295
Mount Eden, 245, 246, 248, 253, 265, Tongue River, 144, 152, 156, 165, 166, Hoh-Kalaloch, 301
267 168 Jasper National Park, 298, 299
Naborton, 160,163 Torrey, 211 Kirchner Marsh, 291
Nenana, 263 Trapper Creek, 245, 246, 248, 252, Kootenai River Valley, 298
Oak Grove (Fork), 144,150,163, 245, 254,265,267 LakeTulane, 283,284
246,248,261, 323 Trout Creek, 60, 100, 101, 102, 126, Llano Estacado, 295
Ocean Point, 245, 248, 264, 265 245, 246, 248, 257, 258, 265 McCollough Range, 295
Paraje Solo, 327 Troutdale, 230 Meadowcroft Rockshelter, 287
Paskapoo, 152,168 Tulelake, 245, 246, 248, 261, 265, 267 Nonconnah Creek, 287
Patterson, 152,153 Upper Cedarville, 212, 232 North Cascades National Park, 298
Perry Place, 152,154 Upper Edmonton, 152,156 Olympic Peninsula, 296
Pinecrest Beds, 246, 248 Upper Ravenscrag, 152 Owens Basin, 296
Piru Gorge, 245, 248,253 Verdi, 245, 248, 253, 267 Potts Mountain Pond, 286
Pitch-Pinnacle, 209 Vermejo, 144, 150, 152, 153, 154,155, Puerto Blanco Mountains, 295
Porcupine River, 263 156 Rita Blanca, 293
Powder River, 144, 155, 194 Vicksburg, 192, 194, 195, 218, 232 Roaring River, 302
Prince Creek, 144, 152, 153, 158 Whitemud, 152, 156, 168 Rogers Lake, 282
Princeton, 149,192, 214, 216, 217, 231 Wilcox, 127, 144, 149, 152, 172, 191, San Agustin Plains, 294
Pudget Group, 209, 217, 230 192,195 San Juan Basin, 294
Puente, 245, 248, 254 age, 170 Santa Barbara Basin, 296
Purple Mountain, 245, 248, 253, 262 Wildcat, 245, 248, 254 Scodie Mountains, 295
Pyramid, 245, 246, 248, 252, 262, Willamette, 212, 230 Sheelar Lake, 284, 287
265,267 Willwood, 144, 150, 165, 166, 167, Sheep Mountain Bog, 298
Raton, lower Raton, 144, 150, 152, 168, 324 Simpsons Flats, 300
153, 154, 155, 160, 161 Wind River, 144, 150, 152, 173, 174, Steens Mountains, 296
Ravenscrag, 144,152,167,168,169, 200, 204, 210, 230 Swan Lake, 299
170 Yakutat/Kushtaka, 144 Tanana Valley Lakes, 303, 304
Redoubt, 217 Yaquina, 222, 230 Tule Springs, 295
Republic, 159, 173, 192, 194, 211, Yellowstone, 144, 150, 152, 174 Tunica Hills, 284
212, 214, 215, 231, 233, 262, 266 York Ranch, 192, 210, 223 Waits Lake, 300
Rex Creek, 217 floras, fossil, Quaternary Washington-Oregon correlations, 301
Ricardo, 245, 248, 253 Alaska correlations, 303, 305 Waterman Mountains, 295
Richards, 192, 222, 232 Aleutian Islands, 302 White Mountains, 296
Ripley, 154, 155 Anderson Pond, 287 Whitewater Draw, 293
Rockdale, 144,150,163 Baffin Island, 302 Wilcox Bluff, 283
Ruby, 60, 104, 192, 200, 210, 223 Banff National Park, 298 Wilcox Playa, 294
Rujada, 192, 221, 222, 232 Beringia, 303 Wolf Creek, 291
San Pablo, 245, 248,254 Big Meadow, 300 Wolsfeld, 291
Scollard, 153 Bonfire Shelter, 294 Yellowstone National Park, 298, 300
Seldovia Point, 194, 212, 229, 233, Boriack Bog, 286 Yoho National Park, 298
254, 262, 266, 267 Buckle's Bog, 286 Florisphera, 40
Sentinel Butte, 144, 150, 152, 163, Camel Lake, 284 Florissantia, fossil, 206, 214, 215, 221,
165, 166, 168, 176 Carolina Bays, 287 222
Sonoma, 245, 246, 248, 254, 265, 266, Carp Lake, 302 Flourensia, 25, 294, 320
267 Chaco Canyon, 295 Fokienia, fossil, 163,167,169,170, 324
Steels Crossing, 212 Clear Lake, 296 foliar physiognomy, 124-128,161, 332
Stewart Valley, 245, 248, 262, 267 Crawford Lake, 288 Forestiera, 135
Stinking Water, 245, 246, 248, 255, Crider's Pond, 289 fossil, 253
258,262,263, 265 Dead Man Lake, 293 formation(s), plant
Succor, Sucker Creek, 100, 104, 122, Death Valley, 295 definition, 3
245, 246, 248, 255, 256, 258, 259, Devil's Lake, 302 of North America, 3
260, 265, 266, 267 Diamond Pond, 298 fossil woods. See dendrochronology
Susanville, 192, 211, 231 East Milford Quarry, 289 Fouquieria, 24, 25
Sutro, 192, 194, 212, 229, 232 Elk Lake, 291 founder effect, 55
344 Index
Fragilaria, fossil, 261
Franseria, 24
Fraxinoipollenites, fossil, 196, 219
Fraxinus, 11, 14, 16, 18, 285, 306, 324,
326
fossil, 107, 168, 174, 194, 197, 199,
202, 210, 220, 226, 228, 246, 249,
251, 252, 253, 255, 256, 258, 259,
260, 262, 283, 284, 286, 288, 289,
290, 291
Fremontodendron, fossil, 211
fronts, weather, 30
Furtunearites, fossil, 213
fusain, 158
Gaillardia, 320
Galium, fossil, 263
Garibaldi-Pemberton belt, 57
Garry a, 21
fossil, 255, 262
Gaultheria, 12,14,15
Gauss-Matuyama boundary, 246, 274
Gaylussacia, 17
genetic distance, 328
geochemical carbon cycle, 43
geoflora(s), concept, 106-109, 149, 306,
322, 331, 332
Cilia, 320
Ginkgo, fossil, 158, 163, 164, 165, 166,
167, 168, 169, 170, 174, 175, 210,
212, 215, 255, 258, 259, 323, 324
glaciations
early, 245
pace, 109
Glacier Peak ash, 298, 300, 302
glacioeustatic, sea-level change(s), 94
Gleditsia, 18, 285, 322
fossil, 250
Gleicheniidites, fossil, 250
Glossotherium, fossil, 104
Glyptostrobus, 255
fossil, 156, 163, 165, 166, 167, 168,
169, 170, 174, 200, 210, 214, 215,
223, 225, 226, 227, 255, 256, 258,
259, 260, 262, 266, 323
Glyptotherium, fossil, 104
Gnetum, 318
Gomphotherium, fossil, 102
Gordonia, 322
fossil, 223, 225, 249, 251
Graminophyllum, fossil, 250
grande coupure, 102,190
Greenland ice cores, 40, 97, 274, 277,
278,279
ice sheets, 245
Grewia type, fossil, 169
Grewiopsis, fossil, 152, 169
Gulf Stream (See also ocean currents),
35
Gunnera, 158
fossil, 175
Gymnocladus, 285, 322 Ilex, 17, 18, 285, 322, 326
fossil, 200, 202, 250, 259, 262 fossil, 121, 163, 174, 194, 195, 197,
gyres, 33, 45 198, 214, 223, 225, 226, 249, 250,
251, 254, 255, 259, 260, 263, 283,
Hadley regime, cell, 30, 33, 58, 142, 286,327
200, 246, 254, 262, 278, 279, 286 Illicium, fossil, 225
Halesia, 285, 322 Inga, fossil, 225
fossil, 206 insolation, 28, 29, 35
Hamamelis, 322 Integricorpus, fossil, 153
fossil, 250, 254, 255 interplanetary dust, 39
Hantkenina, fossil, 189 intertropical convergence zone (ITCZ),
Harmsia, fossil, 170 30,33
Hechtia, 25 Intratriporopollenites, fossil, 164,165,
Hedeoma, 320 196
fossil, 135 lodes, fossil, 213, 214
Heinrich events, 40, 274, 277, 297, lodicrpa, fossil, 213
332 irridium, 66, 160, 189
correlation with terrestrial vegetation, Ischyromys, fossil, 191
283, 284, 296, 302, 304, 331-332 isobars, 30
Heleophyton, fossil, 216 Isoetes, Isoetites, fossil, 166,170, 202,
Hemicyon, fossil, 102,191 249, 250, 283
Heteromeles, 22 isopols, 281
fossil, 228 isostacy, 55
Heuchera, 320 isozyme(s), 320, 328
Hibiscus, fossil, 202 Isthmus of Panama, 35, 246, 278
Hicoria, fossil, 123 /tea, 18
Hillaria, 19 fossil, 215, 283
Hipparion, fossil, 318 Iva, fossil, 249, 283, 287
Hippomaneoidea, fossil, 194
Hippurus, fossil, 220, 226 Jamesia, fossil, 220, 221
Hiraea, fossil, 259 Jarzenipollis, fossil, 197
Hirtella, 122 Javelinoxylon, fossil, 155
Holmesina, fossil, 104 jet stream, 30
Holmskioldia, fossil, 206, 214, 215 Quaternary, 278, 279, 299
Holocene. See Quaternary Joffrea, fossil, 165, 166, 168, 175,
Holodiscus, 13,14 215
fossil, 220, 221, 254, 262 Juglans, 11, 285, 324, 326
Homo, 333 fossil, 107, 121, 123, 152, 163, 194,
Homotherium, fossil, 104 195, 197, 199, 204, 213, 217, 222,
Hooleya, fossil, 199 225, 226, 227, 228, 229, 246, 249,
hopanes, 88, 254, 332 250, 251, 253, 255, 256, 259, 260,
Hordeum, 20 261, 262, 263, 281, 286, 288, 290,
Horniella, fossil, 225 297, 323, 327
horse latitudes, 30 Juncus, fossil, 263
human introduction, migration, 280, Jung layer, 58
281-282, 287, 295, 324 Juniperus, 11, 13, 22, 219, 286, 297,
Humboldt Current (See also ocean 319,320
currents), 31, 33, 254 fossil, 210, 220, 221, 228, 249, 250,
Hydrangea, fossil, 206, 213, 214, 215, 256, 259, 266, 283, 284, 285, 286,
222, 228, 256, 259 292, 293, 294, 295, 296, 297, 298,
hydrogen isostopes, in dendrochronol- 299
ogy, 129 Jussiaea, fossil, 225
hydrological cycle, 28
Hylocomium, 7, 10 Kalmia, 12,17
Hymenoxys, 320 Kandelia, fossil, 174
hypercanes, 160 kaolinite, 217, 242
Hypericum, fossil, 227 Kardiasperma, fossil, 213
Hypertragulus, fossil, 191 Karroo series, 60
Hypohippus, fossil, 100 Keratosperma, fossil, 216
hypsithermal. See postglacial thermal Keteleeria, fossil, 255, 259
maximum Kleinia, 322
 Index 345

Knema, fossil, 107 Sonoran, 11 Magnolia, 17, 18, 285, 318, 322, 323,
Koeberlinia, 25, 320 subalpine, 11 326,328
Koeleria, 19 transition, 11 fossil, 107, 123, 132, 155, 164, 168,
Koelreuteria, fossil, 202, 206, 215, 323 light intensities, for Arctic paleofloras, 175, 213, 214, 225, 229, 249, 254,
Krameria, 295 158 256, 259,323
Liliacidites, fossil, 196 Mahonia, 322
Laevigatosporites, fossil, 196 Limacia, fossil, 173, 174 fossil, 202, 204, 206, 210, 220, 221,
Lagenaria, 287 Limnobiophyllum, fossil, 168 222,228, 251,254,259,260
lag time, 293 Lindera, 18, 120, 323 Malanorrhoea, fossil, 107, 174
Laguncularia, 25, 195, 229 fossil, 164, 202, 206, 213 Malosma, 22, 306
lake effect Lineatum, fossil, 206 Malpoenna, fossil, 107
Eocene, 174, 204 Linum, 120 Mains, fossil, 204, 206, 262
modem, 33 Liquidambar, 17, 18, 285, 319, 323, 325, Malva, 120
Lanagiopollis (Alangium), 211 326,328 Malvastrum, 320
land bridges (See also Beringia, Isthmus fossil, 164, 202, 206, 215, 217, 225, Manilkara, fossil, 197, 223, 225
of Panama, North Atlantic land 226, 229, 246, 249, 250, 251, 255, Manson crater, 68,160
bridge), 60-64, 101, 104, 146-147 259, 263,283, 327 map(s)
Langeria, fossil, 215 Liriodendron, 16, 322, 323 Acer, eastern North America, 325
Langtonia, fossil, 213, 214 fossil, 152, 200, 216, 227, 250, 255, California fault system, 51
lapse rate (See also paleoelevations), 283 Carya, eastern North America, eastern
207, 208 Listera, fossil, 265 Asia, 322
Laramide orogeny, 46, 187, 200, 242 Lithocarpus, 13, 15, 123 central Rocky Mountains, Paleocene,
Larix, 8, 9, 10, 15, 16 fossil, 210, 229, 254, 259, 262 early Eocene, 173
fossil, 174, 199, 204, 211, 214, 223, Litsea, 120 continental positions
226, 227, 228, 229, 233, 249, 250, fossil, 173 Cretaceous, 145
251, 259, 262, 263, 264, 265, 284, little ice age, 60, 279, 281, 291, 296, 299 Eocene, 147
285, 290, 291, 300,326 Loiseleuria, 7 Paleocene, 146
Larrea, 19, 24, 25, 294, 295, 320, 321 Lomatia, 202, 206 early Paleocene vegetation, 154
fossil, 295 Long Valley caldera, 57 Eocene coast line, 148
Lassen Peak, 57 Lonicera, fossil, 199, 211 Eocene lakes, western United States,
Laurentide ice sheet, 245, 275, 276, 278, Loranthus, fossil, 196 201
280, 287, 289, 290, 292, 299, 301 Ludwigia, fossil, 197, 227, 251, 263 extent of glaciation, 275
Laurocalyx, fossil, 213 Luehea, fossil, 173 Fagus, eastern North America, 325
Laurocarpum, fossil, 213 Luvunga, fossil, 107, 174 Juniperus, 297
Laurocerasus, fossil, 253 Luzula, 7 late middle Eocene vegetation, 195
Laurus, fossil, 152 Lycopodium, 10, 322 Liquidambar, eastern North America,
Lavatera, 319, 320 fossil, 168, 196, 222, 226, 249, 251, 325
leaf margin analysis (LMA), 128 259,264,265 North American deserts, 23
leaf physiognomy. See foliar physiog- Lycopodiumsporites, fossil, 196 Nyssa, eastern North America, 325
nomy Lycopus, fossil, 226 physiographic features, 47, 49, 151,
leaf-size index, 124-125 Lygodium, fossil, 166, 167, 170, 195, 153, 193, 247
Cretaceous, 152,160 196, 200, 202 Pinus edulis, modern, late Wisconsin,
lower Atanikerdluk flora, 155 Lyonia, 17, 18 294
Puget Group, 217 fossil, 225 place names, 48-49
Ledum, 7 Lyonothamnus, fossil, 228, 262 Pleistocene lakes, Great Basin, 292
Leguminocarpum, fossil, 213 Lysichiton, fossil, 302 vegetation, 4
Leguminosites, fossil, 202, 206 Lythrum, fossil, 216 Maranta, 122
Leitneria, fossil, 283 Mastixia, fossil, 213, 214, 323
Leptarctus, fossil, 191 Macginicarpa, fossil, 168, 212, 213, 215 Mastixicarpum, fossil, 213, 215
Leptomeryx, fossil, 102, 191 Macginitiea, fossil, 168, 169, 212, 215 Matuyama-Brunhes reversal, 275
Leucophyllum, 25 Machilus, fossil, 173, 221 Matuyama chron, 264
Leucothoe, 18 Madura, 285 mean annual range of temperature
Libocedrus, 13 fossil, 293 (MART)
fossil, 259 Macrocoma, 327 Eocene-Oligocene, 229, 230
life zones Macromitrium, 327 late Miocene, 252
alpine, 11 Madrean mean annual temperature (MAT)
Arctic, 11 Madrean-Tethyan hypothesis, 319 early Miocene, Miocene, 187, 230
Canadian, 11,13 region, 11, 219 Eocene, 187, 230
foothill, 11 vegetation, 211, 262, 266, 319, 320 glacial maximum, 283
Hudsonian, Maesa, fossil, 173 land, 33
montane, 11, 12, 13 magnetostratigraphy, 98 middle Miocene-Pliocene, 267
346 Index
mean annual temperature (MAT)
(continued)
ocean water, 33
Oligocene, 187, 230
summary, middle Eocene through
early Miocene, 231
Mediterranean climate, 21, 53, 306
definition, 12
origin, 253-254, 321
Melia, fossil, 217
Meliosma, fossil, 165,166,167,169,
174, 213,214, 215,251,252
Menispermites, fossil, 165, 167, 202
Menispermum, fossil, 165
Menocems, fossil, 191
Mentzelia, 320
Menyanthes, 7
fossil, 226, 227, 265
Menziesia, 15
Merychippus, fossil, 100
Merycoidodon, fossil, 191
Mesocyparis, fossil, 170
Mesohippus, fossil, 102,191
Messinian salinity crisis, 96,104,
242
Metasequoia, 158
fossil, 156, 157, 163, 165, 166, 167,
168, 169, 170, 174, 175, 196, 199,
200, 204, 210, 215, 216, 217, 221,
222, 226, 227, 254, 255, 256, 259,
323,324
Meyrcoidodon, fossil, 102
Micrhystridium, fossil, 259
Microdiptera, fossil, 223, 227
Microtus, fossil, 294
middle Eocene diversity decline
(MEDD), 197, 199, 211
middle Holocene climatic optimum, see
postglacial climatic optimum
middle Pliocene warmth, 242, 246
Milankovitch variations, 36-40, 200,
245
Milfordia, fossil, 196, 211, 219
Millettia, fossil, 173
Mimosa, 25
fossil, 135
Mimosites, fossil, 202
Mimusops, fossil, 225
Miocene thermal optimum, 188
Miomastodon, fossil, 102
Mionictis, fossil, 191
Missoula floods, 68-70
Mneme, fossil, 227, 249, 250
modern analog method, 105,123-124,
207
Mohria, fossil, 194, 195, 196
Mohrodendron, fossil, 107
molecular clock(s), 320, 328-329
Momipites, fossil, 164, 168, 174, 194,
197, 198,219, 251
Monocolpopollenites, fossil, 196
Monocotylophyllum, fossil, 221
monte, 320
Monttea, 320
Moropus, fossil, 191
Moms, 285, 323
fossil, 206, 222, 223, 225, 227, 251
Mount Mazama, ash, 57, 70, 281
Mount St. Helens, ash, 57, 58, 70, 257,
281, 298, 300
mountain systems, origin, 46-52
Mt. McKinley, 52
Muhlenbergia, 19
Musa, fossil, 214
Musophyllum, fossil, 163, 169, 175, 200
Musopsis, fossil, 163, 169
Myexiophyllum, fossil, 168
Myrica, 285, 326
fossil, 121, 152, 168, 195, 197, 210,
211,227, 249,251,286
Myricophyllum, fossil, 170
Myriophyllum, fossil, 197, 251, 265
Myristica, fossil, 107, 174
Myrtaceidites, fossil, 197, 219
Myrtillocactus, 25
Myrtophyllum, fossil, 157
Myrtus, fossil, 195,197
Mytonolagus, fossil, 190
Nassella, fossil, 250
nearest living relative method (NLR),
123
Nectandra, 120
fossil, 152, 214, 221
Nelumbago, fossil, 170
Nelumbo, fossil, 166
Nemaleison, fossil, 167
Neocallitropis, fossil, 157, 175
Neogenisporis, fossil, 251
Neolitsea, fossil, 173
Neotoma (See also pack rat middens),
115,131,133,134
Nestronia, fossil, 225
Newark Basin, 38
Nipa (Nypa. See also Proxapertites),
fossil, 98, 162, 191, 194, 195, 229,
266
Nitella, fossil, 226
nobel gases, 285
Molina, 320
Nordenskioldia, fossil, 165,168,175,
196, 215,255,323
Normapolles, 150, 153, 156, 162
North American Land Mammal Age(s)
(NALMA), 94, 98-106, 190
in relation to fossil floras, 104
unsettled age boundaries, 190
North Atlantic Deep Water (NADW),
34-35,41, 187
North Atlantic land bridge, seaways,
61-64, 187, 191, 199
notophyllous, polar broad-leaved
evergreen, deciduous forest, 149,
150, 153, 156, 158, 161, 165, 170,
174, 175, 176, 177, 194, 195, 217,
331
Notrotheriops, fossil, 280
nunatak hypothesis, 318
Nuphar, fossil, 215, 220, 226, 250
Nuxpollenites, fossil, 194, 197
Nymphaea, fossil, 226, 258, 259, 262
Nymphaeites, fossil, 228, 258
Nymphoides, fossil, 226
Nyssa, 17, 18, 285, 322, 323, 325, 326
fossil, 174, 194, 195, 197, 198, 200,
210, 213, 214, 223, 225, 229, 249,
250, 251, 255, 259, 260, 262, 263,
266, 281, 283, 286
Nyssidium, fossil, 165, 166, 175, 323
ocean currents, 33, 34, 45
gradients, 187
Ochroma, fossil, 202
Ocotea, fossil, 194, 202, 204, 214, 221,
225,233
Odontocarya, 122
Odontocaryoidea, fossil, 213
Oenothera, fossil, 263
Oligocene deteriortion, 189,190
Olneya, 320
fossil, 320
Olympic Peninsula, 14
Onoclea, fossil, 108, 168, 170
Ontung Java Plateau, 142, 144
Onychomys, 320
Oplopanax, 15
Opuntia, 19, 25, 295
fossil, 135
Oreomunnea, fossil, 163, 164, 198, 249
Oreopanax, fossil, 173, 202, 206, 256,
258, 259, 262
Oreora, fossil, 194
orogeny, 30, 44-56
effects on climate, 44, 52-56, 187
effects on vegetation, 54-55
Osmanthus, fossil, 202, 206
Osmaronia, fossil, 221
Osmunda, 10
fossil, 166, 168, 170, 174, 216, 222,
226, 249, 251, 259, 264
Ostrya, 121, 285, 326
fossil, 194, 204, 226, 249, 250, 252,
255, 259, 261, 283, 284, 288, 289,
290, 291
otoliths, 88, 261, 332
Ovoides, fossil, 259
Oxydendrum, 285
fossil, 225
oxygen isotope ratios, 37-38
Caribbean, Indian Ocean cores, 276
early Oligocene, 190
for determining paleotemperatures,
88

from Antarctic current, 187
in tooth enamel, 43
Quaternary, 277
Pachycormus, 320
Pachysandm, fossil, 168,174, 211, 217,
259, 323
pack rat middens, 115,128,131-137,
256, 293, 294, 295, 332
pollen and spore content, 133,137,
281
Palaeoallophylus, fossil, 213
Palaeolagus, fossil, 191
Palaeophytocrene, fossil, 165, 213,
215
Palaeosinomenium, fossil, 213
Paleocarpinus, fossil, 164,165,168,
215,323
Paleocarya, fossil, 199, 213
paleoelevations (See also lapse rates),
55, 56, 92, 128, 172, 187, 207-210,
214,252, 332
paleofloristics, 148
paleomonographic studies, 148-149,
196,332
Paleomyrtinaea, fossil, 165, 215, 216
Paleopanax, fossil, 213
Paleoplatycarya, fossil, 213
Paleorosa, fossil, 216
paleosols, pedotypes, 42, 332
early Eocene, 170,198
K-T boundary, 161
late Miocene-Pliocene, 250, 263
Maastrichtian, 155, 160
middle Eocene-early Miocene, 189,
199,211, 226
Oligocene, 218-219
paleotemperature(s), climate 86-92, 327
abrupt changes, pulses, pauses (See
also vegetation), 166-167,
189-190, 198, 245, 296, 306,
332
Cretaceous, 147
curve, 89
Dorf curve, 89-91
Eocene (thermal) maximum, 147
gradients, see thermal gradients
hemisphere correlations, 296,
302-303, 306
Paleocene, Ft. Union, Willwood,
167
Quaternary, 278, 279
re evolution, 198
terrestrial vegetation, 89-92
Wolfe-Hopkins curve, 91
Paliurus, fossil, 173, 227
palynology. See pollen and spores
Panicum, 19
fossil, 250
Parabombacaceaeoxylon, fossil, 157
Paracarpinus, fossil, 206, 221
Parana series, 60
Pamnymphaea, fossil, 170
Paraoreomunnea, fossil, 194
Paraphyllanthoxylon, fossil, 132
Parashorea, fossil, 174
Parataxodium, fossil, 157,165,166,175
Paraternstroemia, fossil, 166
paratropical (rain) forest, 149,153,154,
155, 158, 160, 165, 170, 174, 175,
176, 177, 194, 195, 199, 200, 211,
217,331
Paris, 322
Parthenocissus, fossil, 197, 206, 212,
213,250,255
Parvileguminophyllum, fossil, 194, 202
Paulownia, fossil, 255
Paxistima, fossil, 204
Pediastrum, fossil, 259, 264, 280
Penosphyllum, fossil, 167
Pentoperculum, fossil, 213
Peraphyllum, fossil, 220
Peridinium, fossil, 225
periglacial, 278
perihelion, 29, 39
Perityle, fossil, 135
permafrost
Arctic, 245
definition, 3
Quaternary, 276, 287, 297, 303
Persea, 18, 120, 285, 323
fossil, 107, 164, 165, 168, 202, 206,
210, 213, 225, 253, 254, 259, 266
Persites, fossil, 167
Petaurista, fossil, 104
Petrea, fossil, 206
Peucephyllum, 295
fossil, 295
Phaca, fossil, 206
Phacelia, fossil, 295
Phaseolites, fossil, 206
Phaseolus, fossil, 287
Phellodendron, fossil, 225
Phenacodus, fossil, 167
Philadelphus, 15
fossil, 206, 220, 253, 262
Philodendron, fossil, 194
Phleum, 12
Phoebe, fossil, 167, 215, 254
Phoenix, 120
photic zone, 88
Photinia, fossil, 215, 254, 259, 262
photoperiod, 29
Phyllanthus, fossil, 226, 227
Phyllodoce, 7,13
Physalis, fossil, 227
Physocarpus, fossil, 220
Phytocrene, fossil, 174
Picea, 3, 7, 8, 9,10,11,12,14,15,16,
55,120, 159, 195,282
fossil, 121,168, 174, 195, 196, 199,
200, 204, 205, 210, 211, 214, 215,
Index 347
220, 221, 222, 223, 226, 227, 228,
229, 231, 233, 249, 251, 252, 254,
255, 256, 259, 260, 262, 263, 264,
265, 266, 268, 281, 284, 285, 286,
287, 289, 290, 291, 292, 293, 295,
297, 298, 299, 300, 301, 302, 303,
304, 305, 306, 326, 327
piedmont, 1
Pieris, 322
pine barrens, 17
Pinus, 8, 10,11, 12, 13, 14,15, 16,17,
18, 22, 55, 120, 129, 130, 131, 195,
219, 282, 295, 296, 306, 318, 319,
320,326
fossil, 121, 128, 168, 174, 194, 195,
196, 198, 199, 200, 202, 204, 205,
206, 211, 212, 213, 214, 215, 216,
219, 220, 221, 222, 223, 225, 226,
227, 228, 229, 233, 246, 249, 250,
251, 252, 253, 254, 255, 256, 259,
260, 261, 262, 263, 264, 265, 266,
281, 283, 284, 285, 286, 287, 288,
289, 290, 291, 292, 293, 294, 295,
296, 297, 298, 299, 300, 301, 302,
304, 305, 306, 324
Pistia, fossil, 170
Pistillipollenites, fossil, 170,197
Pityrogramma, 169
Plafkeria, fossil, 222
Plagiopodopsis, fossil, 205, 206, 208
Planera, fossil, 107
Plantae Rariores Camschatcenses, 322,
323
Plantago, fossil, 265
Platananthus, fossil, 168, 212
Platanites, fossil, 167,170
Platanus, 11, 255, 285, 319, 320, 326
fossil, 163, 164, 165, 166, 167, 168,
169, 194, 195, 197, 199, 200, 202,
203, 206, 213, 214, 217, 222, 229,
249, 250, 251, 252, 253, 254, 255,
258,259,283,284,323,324
plates, North America, 47, 50
Platycarya, fossil, 107, 163, 164,165,
167, 170, 172, 173, 174, 194, 200,
323
playas, 281, 293, 284
Pleistocene. See Quaternary
Plesiosminthus, fossil, 191
Pleurozium, 10
pluvial climates, intevals, lakes, 278,
292, 294, 301, 320
Poa, 12, 20
Podagrostis, 12
Podocarpus, 195
fossil, 170, 194,195,196, 198, 249,
251,259
Poebrotherium, fossil, 102,191
Polanisia, fossil, 227
polar broad-leaved deciduous forest.
See notophyllous
348 Index
polar broad-leaved evergreen forest. See
notophyllous
polar inclination, Arctic fossil floras,
158-159
pollen and spores, 115-121
absolute pollen frequency, 282
in metamorphic rocks, 141
in meteorites, 141
in pack rat middens, 133,137
methodologies, Quaternary,
281-282
moss polsters, 282
pollen influx, 282, 296
preservation, Quaternary, 280
Pollenites, fossil, 197
Polygonella, fossil, 283
Polygonum, 320
fossil, 249, 286
Polypodium, fossil, 222, 249, 259
Polyptera, fossil, 163, 167
Polytrichum, 7
Populus, 8, 10, 11, 12, 13, 14, 16, 18,
117, 255,285
fossil, 107, 167, 173, 199, 200, 202,
203, 204, 206, 214, 215, 220, 221,
223, 226, 228, 233, 250, 251, 252,
253, 254, 255, 258, 259, 260, 261,
262, 263, 264, 265, 281, 286, 288,
290, 293, 303, 304, 305, 327
Porosia, fossil, 165,168
Portulaca, fossil, 288
postglacial climatic optimum, 279, 280,
287, 291, 292, 294, 295-296, 297,
298, 299, 302, 306
Potamogeton, fossil, 152, 167, 202, 226,
227, 250, 259, 265
Potentilla, 7
fossil, 210, 215, 221, 227
Potomotherium, fossil, 191
prairie, 19
prairie peninsula, 19
precipitation
extremes, 35, 36
Mio-Pliocene aridity, 250, 262
trends, Eocene, Rocky Mountains,
209
preservation, fossil floras, 99-100, 187,
189
grasses, 204
Princetonia, fossil, 214, 216, 217
Proboscidea, 320
Prosopis, 19, 22, 24, 25, 286, 320
fossil, 206, 210, 284, 285, 295
Proteoides, fossil, 155,175
Protoceras, fossil, 191
Protomimosoidea, fossil, 170, 172
Protungulatum, fossil, 156
Proxapertites (See also Nipa], fossil,
162,194,196
Prunites, fossil, 170
Prunus, 13, 14, 22, 285, 319, 320, 326
fossil, 123, 199, 202, 206, 210, 213,
215, 216, 217, 220, 221, 226, 228,
250, 251, 253, 255, 259, 261, 262
Pseudaelurus, fossil, 102, 191
Pseudolaesopollis, fossil, 197, 251
Pseudolarix, fossil, 168, 174, 204, 214,
215,233, 259,305
Pseudophoenix, fossil, 196
Pseudotheridomys, fossil, 191
Pseudotsuga, 11, 13, 14, 15
fossil, 210, 211, 220, 228, 252, 253,
254, 255, 256, 259, 263, 293, 299,
300, 302
Psidium, fossil, 165
Psophosphaera, fossil, 259
Ptelea, fossil, 123, 202, 206, 250, 259,
262
Pteleaecarpum, fossil, 214, 221, 222
Pteridium, fossil, 226
Pterocarpus, fossil, 223
Pterocarya, fossil, 163, 168, 173, 174,
194, 195, 197, 199, 200, 202, 203,
204, 214, 215, 217, 222, 225, 226,
229, 249, 250, 251, 252, 254, 255,
258, 259, 260, 262, 263, 266, 323,
324
Pteronepelys, fossil, 215
Ptychopetalum, fossil, 196
Puccinellio, 4
pumps, CO2 (see also sinks), 43-44, 93
Purshia, 21
Pyrenancantha, fossil, 174, 213
Pyrus, 285
fossil, 259
Quaternary
beginning, 89, 274
extent of glaciers, 275, 287
ice retreat, 279
ice thickness, 275, 278
pace of glaciations, 109
re Milankovitch variations, 274
stages, subdivisions, 274, 275, 280,
282, 307
Valley Forge, 279
Quercoidites, fossil, 197
Quercus, 11, 13, 14, 15, 16, 17, 18, 20,
21, 22, 120, 128, 149, 255, 285, 295,
296, 306, 319, 320, 326
fossil, 121, 123, 128, 164, 167, 168,
170, 174, 194, 195, 197, 198, 199,
200, 202, 204, 206, 210, 213, 214,
217, 218, 219, 220, 222, 223, 225,
226, 228, 229, 233, 249, 250, 251,
252, 253, 254, 255, 256, 258, 259,
260, 261, 262, 263, 281, 283, 284,
286, 288, 289, 290, 291, 293, 296,
297, 299, 302, 306, 323, 327
Ramorinoa, 320
Ranunculus, fossil, 220, 227
rebound, ice surfaces, 279
redeposition, pollen and spores, 120
refugia, 282, 284, 287, 289, 294, 301,
303,304,324, 328
Republica, fossil, 215
Rhacomitrium, 7
Rhamnites, fossil, 170
Rhamnus, 21, 285, 319, 320
fossil, 107, 152, 204, 223, 225, 253,
296
Rhizophora, 25, 26, 121, 177, 195, 229,
266, 283, 306
Rhododendron, 7, 14, 15, 123
fossil, 107, 204, 210, 225, 254
Rhoipites, fossil, 197
Rhoiptelea, fossil, 121
Rhus, 21, 22, 306, 319, 322, 326
fossil, 123, 197, 202, 206, 213, 215,
225, 253, 255, 259, 262, 263, 286,
296
Ribes, 13
fossil, 206, 211, 215, 216, 219, 220,
221, 226, 228, 251, 259, 262, 293
Robinia, 322, 323
fossil, 199, 206, 220, 226, 228, 251,
262
Rocky Mountains, 46-47, 55, 142, 146,
172, 173, 187, 200, 204, 242
Rorippides, fossil, 170
Rosa, fossil, 202, 206, 220, 228, 255,
262
Rossby regime, 30, 200, 246
Rubus, 7, 15
fossil, 220, 225, 227
Ruellia, fossil, 211
Rumex, fossil, 227, 287, 288, 289
Ruptiliocarpon, 109
Sabal, 18, 285, 323
fossil, 163, 164, 194, 195, 196, 213,
229, 266
Sabia, fossil, 213, 214, 215
Sageretia, fossil, 174
Sagisma, fossil, 227
Salicornia, 23
Salina terrane, 64
salinity differences, concentrations, 33,
34
Quaternary, 276-277
Salix, 7, 8, 10, 11, 13, 16, 18, 285
fossil, 123, 152, 200, 202, 204,
206,208,210,215,220,221,
223, 226, 228, 249,250,251,
252, 253, 254, 255, 259, 260,
261, 262, 263, 264, 265, 281,
283, 286, 289, 290, 291, 293,
298, 302, 303, 304, 305
Salixipollenites, fossil, 197
Salton Sink, 253
Salvia, 21, 320
fossil, 296
 Index 349

Salvinia, fossil, 166, 167 Sida, 320 Suaeda, 23

Sambucus, 285 sidewalk hypothesis, 318 subduction, rates, 57, 60
fossil, 206, 220, 227, 250, 297 Sierra Nevada, 50, 51, 52, 142, 187, 204, Subhyracodon, fossil, 191
San Andreas fault, 51 242 subtropical forest, 149-150
Sanicula, 306 Sigmapollis, fossil, 259 succession, 306
Santa Ana winds, 33 Silene, 7 Swartzia, fossil, 202
Sapindus, 323 fossil, 291 Symphoricarpos, 13, 21, 322
fossil, 170, 200, 202, 206, 220, 250, silicate rocks, erosion re CO2 concentra- fossil, 220, 228, 256, 259, 263
253,262,293 tion, 44, 93, 187, 189, 278 Symplocos, 322
Sapium, 24 Siltaria, fossil, 197, 219, 223, 225 fossil, 162, 194, 197, 202, 213, 223,
fossil, 214, 295 sinks (CO2), 43-44, 93 251,255, 323
Sarcobatus, 23 Siparuna, fossil, 214
fossil, 251, 252, 259, 260-261, 268, Smilax, 326 Tanyoplatanus, fossil, 213
296, 298, 299 fossil, 205, 214, 215, 250, 259 Tapeinochilos, 117
Sarcococca, fossil, 211 Solarium, fossil, 227 taphonomy, 141
Sargentodoxa, fossil, 109, 223 solar constant, 35 Tapiscia, fossil, 213, 323
Sassafras, 120, 285, 322, 323 solar flares, 35 Tarpon, fossil, 98
fossil, 164, 206, 211, 215, 254, 255, Soleredera, fossil, 216 Taxodiapollenites, fossil, 156
256,258,259,262 solstice, 29 Taxodium, 18, 129, 323
Satureja, 306 Sophora, fossil, 249, 250, 255, 262 fossil, 158, 167, 168, 170, 174, 175,
Saurauia, fossil, 174 Sorbus, 15 199, 210, 214, 227, 229, 249, 250,
Saxifraga, 4, 7 fossil, 220, 221, 222, 228 252, 254, 255, 256, 258, 259, 260,
Saxifragispermum, fossil, 213 Sorghastrum, 19 266,283
scablands, 68-69, 302 South Atlantic Deep Water (SADW), taxol, 14
Schinopsis, 121 34 Taxus, 14, 15
Schisandra, fossil, 213, 214 southern oscillation, see El Nino fossil, 212, 213, 215, 249
Schistidiium, fossil, 249-250 Sparganium, fossil, 165,167,168, 202, TCT, fossil, 249, 251, 259, 261, 263,
Schizaea, fossil, 168 219,226,227, 251 296
Schkuhria, 320 Spartina, 19 telluric changes, 44
Schmaltzia, fossil, 206 specific heat Teloceras, fossil, 104,191
Schoepfia, fossil, 167, 215 land, 33, 93 temperature(s)
schwingpolen hypothesis, 318 oceans, 33, 93 CLIMAP estimates, marine, 86
Sciadopitys, fossil, 249, 251, 263 Sphagnum, 7, 8, 10, 12 extremes, 35, 36
sea breezes, 33 fossil, 163, 168, 174, 196, 222, 226, ranges, re vegetation, 125
sea-level changes, 89, 92-98 251, 264, 265 tephra, tephrachronology, 98, 281
Cretaceous, 144 Spinozonocolpites, fossil, 191 terminal Eocene event, 189,190,
curve, 95, 97 Spiraea, 322 197
middle Eocene-early Miocene, 188 fossil, 215 Terminalia, fossil, 107
Neogene, 242 Spirematospermum, fossil, 157,165,175 terminations (glacial stages), 39, 40,
Oligocene, 190 Spirodela, fossil, 167, 170 274,276,280
Quaternary, 276 Sporobolus, 19 Ternstroemites, fossil, 165, 215
re land bridges, 64 Staphylea, 319 terranes, 64-65,107
re temperatures, 96 fossil, 206 Alaska, 174, 175
seasons, 28-29 Stemonurus, fossil, 174 Pacific Northwest, 170
Sedum, fossil, 227 Stenocereus, fossil, 295 Tethys Sea, 319
Selaginella, fossil, 135, 196, 226, 250, Stephanodiscus, fossil, 261 Tetracentron, 165,175
251, 264,265,283 steppe, 295, 298, 301, 324 Tetracolporopollenites, fossil,
Semecarpus, fossil, 107 definition, 23 197
Sequoia, 14, 15, 158, 322 Sterculia, fossil, 202 Tetradymia, 23
fossil 157, 167, 174, 175, 196, 202, sterenes, 88, 254, 332 Teucrium, fossil, 227
204, 205, 206, 208, 210, 215, 222, Stereocaulon, 10 Thamnolia, 7
249, 251, 254, 255, 259, 266 Stewartia, 285 Thanikaimonia, fossil, 213
Sequoiadendron, 7,14,158 Stipa, 19, 23 Thelypteris, fossil, 167, 200
fossil, 255 fossil, 205 thermal contraction, expansion (sea
Sequoiapollenites, fossil, 156 Stockeya, fossil, 221 water), 96,97
Serenoa, 17 stomata, estimates of ploidy level, thermal gradient(s)
fossil, 195, 196, 229 256 early Eocene, 169
Serjania, fossil, 173 stomatal analysis, index, 128, 332 Maastrichtian, 157,199
Sesuvium, fossil, 227 Striopollenites, fossil, 197, 219 middle Eocene, 199
Shepherdia, fossil, 220, 259, 261, 291, Styrax, 18, 285, 319, 320 thermohaline cell, circulation, 35, 40,
300 fossil, 202 278, 296, 297, 302
350 Index
Thouinia, fossil, 173, 202, 206
Thuja, 8, 10, 14, 15, 323
fossil, 204, 210, 211, 214, 215, 226,
227, 233, 249, 255, 259, 266, 285,
288, 302, 305, 326, 327
Thulean
route, 63,101,147,172
Volcanism, 63,170
Tibetan Plateau, 52-53, 55
Ticholeptus, fossil, 100
Tilia, 16, 285, 324, 326
fossil, 168, 174, 194, 195, 197, 199,
200, 206, 214, 215, 222, 225, 231,
249, 251, 259, 262, 263, 265, 266,
283, 286, 288, 289, 291
time scale
Cretaceous (Maastrichtian),
Paleocene, early Eocene, 143
early Miocene, 224
middle late Eocene, 171
middle late Miocene, 243
Oligocene, 218
Paleogene, Gulf Coast, 172
Pliocene, 244
Tinomiscoidea, fossil, 213
Tinospora, fossil, 213
Toba, eruption, 57
Todda, fossil, 225
Tomenthypnum, 7
Torreya, fossil, 205, 213, 215
Toxicodendron, 285
fossil, 202, 250
trade winds, 30, 58
Trapa, fossil, 249, 250, 258, 323, 324
Trapago, fossil, 170
tree-ring analysis. See dendrochronolgy
Trema, fossil, 213
Tremophyllum, fossil, 222
Triatriopollenites, fossil, 211
Triceratops, fossil, 145,156
Trichilia, fossil, 206, 210
Trichocereus, 320
Tricolpites, fossil, 158
Tricolporopollenites, fossil, 197
Triprojectacites, fossil, 150,162
Triumfetta, fossil, 202, 211
Trochodendroides, fossil, 166,167,169,
170, 323
Trochodendron, 165, 175
fossil, 215, 255, 323
Tropic of Cancer, 29
Tropic of Capricorn, 29
tropical dry, semideciduous, forest,
191, 195, 198,204,211,229
tropical forest, 149,152,153,165,
175, 176,229, 331
tropical rain forest, 149,154,162,165,
169, 170, 171, 176, 177, 195, 211,
212,223,229, 331
Trudopollis, fossil, 150, 153
Tsuga, 10, 13, 14, 15, 16, 282, 323
fossil, 195, 196, 198, 199, 204, 211,
214, 215, 220, 222, 223, 226, 227,
228, 229, 231, 233, 249, 251, 252,
254, 255, 259, 260, 262, 263, 264,
266, 268, 288, 289, 291, 296, 300,
302,303, 305
Tsukada, fossil, 215
Tubela, fossil, 226
tundra
Arctic, early, 226, 249-250, 266
Cascade Mountains, 12
Quaternary, 286, 287, 290, 303, 304,
306
Sierra Nevada, 12
tunguska (near-hit asteroid), 68
Turpinia, fossil, 223
Typha, fossil, 200, 202, 205, 206,
210, 219, 228, 250, 251, 259,
260,261
Uhlea, fossil, 216
Ulminpollenites, fossil, 164
Ulmus, 16, 18, 255, 285, 288, 324,
326
fossil, 107, 121, 126, 164, 167, 168,
174, 194, 195, 197, 199, 206, 210,
211, 215, 217, 222, 223, 225, 226,
228, 246, 249, 250, 251, 252, 254,
255, 256, 258, 259, 260, 261, 262,
263, 266, 281, 283, 284, 286, 288,
289,290, 291, 323,327
Umbellularia, 15, 22,120
fossil, 254, 259
Ungnadia, fossil, 173
upwelling, 33-34, 246, 262
Ursavus, fossil, 102,191
Ursus, fossil, 104
Uvularia, 322
Vaccinium, 7, 12, 13, 15, 285
fossil, 220, 225, 226, 228, 254, 259,
260, 263, 291
Valeriana, fossil, 265
varves, 281
Vauquelinia, fossil, 202, 206, 295
vegetation
definition, 3
early decline exotic elements, 200,
204
late decline exotic elements, 242, 249,
252, 262, 265, 268
map, North America, 4
step-wise changes (See also paleo-
temperatures) 197-198, 211, 226,
230,245
ventilation rate, 38
Verbena, fossil, 227
Verrucatosporites, fossil, 196
Verrutricolporites, fossil, 197
vertical sesimic profile, 94
Viburnum, 285
fossil, 158, 164, 165, 168, 170, 174,
175, 196, 199, 204, 214
vicariance, 55, 94, 330
biogeography, 318, 329, 330, 331
Vicia, fossil, 206
Viguiera, 295
fossil, 135
Viola, fossil, 226
Vitaceoxylon, fossil, 214
Vitis, 322
fossil, 206, 213, 215, 223, 225, 226,
250,251,283
volcanism, volcanoes, 56-60
effect on vegetion, fossilization, 60,
100,255,256-257
re sea-level changes, 96
Wallace's line, 44
warm interglacials, as atypical, 280,
287, 306
warm temperate forest, 150
water vapor, 28
Wehrwolfia, fossil, 216
Weigela, fossil, 227
westerlies (winds), 30
Woodwardia, fossil, 167,168,170,
200, 259
Wrangell Mountains, Wrangellia,
64
Xerophyllum, 15
Younger Dryas, 40, 41, 277, 278, 279,
288, 290, 292, 296, 302, 332
Yucca, 19, 24, 25, 295, 320
fossil, 135, 295
Zanthoxylum, 285, 322
fossil, 197, 223
Zea, 122
fossil, 287, 288
Zelkova, fossil, 121, 164, 174, 197,
199, 200, 202, 203, 206, 210,
223, 226, 228, 229, 249, 250,
251, 252, 254, 255, 256, 259,
260, 262, 323, 324, 332
Zenobia, fossil 225
Zingiberopsis, fossil, 157,166,167,
175,200
Zizyphoides, fossil, 255, 323
Zizyphus, 320
fossil, 152, 206, 220
zonation, see life zones
zooxanthellate (corals), 34
Zuccagnia, 320